sign in sign up
<<
Home , Banal nationalism, East Turkestan, Ethnotaxonomy, Great Mosque of Hohhot, Han Chinese nationalism, Han nationalism

NATION-BUILDING AND ETHNIC BOUNDARIES IN ’S NORTHWEST

A thesis submitted to The University of Manchester for the degree of

Doctor of (PhD)

in the Faculty of Humanities

2013

David Tobin

SCHOOL OF SOCIAL SCIENCES /Department of Politics

Table of Contents

List of Figures ...... 5 List of Abbreviations ...... 6 ABSTRACT ...... 7 Declaration and Copyright ...... 8 Acknowledgements ...... 9 Dedication ...... 11 The Author ...... 12 Introduction ...... 13 Who is ‘China’ and who is this ‘we’, which ‘we’ want to secure? ...... 15 Power and Resistance in Textual ...... 18 Research Questions:...... 22 Thesis Outline: -Building and Ethnic Boundaries ...... 23 Chapter 1: Securing the Nation in a Multi-Ethnic “Frontier” ...... 27 Section 1: Security and Identity in International Relations Theory ...... 28 1.1: De-centring Security Studies ...... 28 1.2: Performativity and the Production of Boundaries...... 30 Section 2: and Nation-Building ...... 33 2.1 and Nation-ness ...... 34 2.2: Nation-Building in and of ‘China’ ...... 38 Section 3: Ethnicising Boundaries from Above and Below ...... 49 3.1: Top-Down Approaches: Ethnicity as Ascription ...... 51 3.2: Bottom-Up Approaches: Ethnicity as Self-Identification ...... 55 Section 4: Methodological Challenges From Above and Below ...... 58 4.1: A Textual approach: Discourse and the Social Production of ...... 59 4.2: An Ethnographic Approach: Addressing the Question of Discourse’s Address...... 61 Conclusions ...... 65 Chapter 2: Writing the Nation: Performing a Timeless Self/Other ...... 67 Introduction ...... 67 2.1: Unifying Minzu in the Party-State’s History of , China, and the World ...... 70 2.2: Producing Minzu-ness, Producing Han-ness ...... 79 2.3: Overcoming Minzu with Han-ness ...... 84 Conclusions ...... 94

2

Chapter 3: East and the Production of the Internal/External Boundaries of - ness ...... 97 Introduction ...... 97 3.1: and the Securitisation of Internal/External Boundaries ...... 99 3.2: “The Three Evils” and the Production of Resistance ...... 108 3.3: The Securitisation of Modernisation: Boundaries within Boundaries within Boundaries ...... 119 Conclusions ...... 127 Chapter 4: Performing the Securitisation of Uyghur Ethnicity ...... 129 Introduction ...... 129 4.1: Producing Meanings of Violence: Securitisation and the Internal/External Security Dilemma ...... 130 4.2: Producing Meanings of Violence: Nation-Building and the Securitisation of Minzu Boundaries ...... 138 4.3: The City After “7-5”: (In)Securitisation and the Ethnic Threat ...... 150 4.3.1: Syringe Assaults and Everyday (In)Security ...... 152 4.3.2: and Popular ...... 158 Conclusions ...... 163 Chapter 5: Performing Uyghur Inclusion, Objectivising Han-ness ...... 166 Introduction ...... 166 5.1: Hierarchical Inclusion/Exclusion and the Struggle for Minzu Tuanjie ...... 167 5.2: Ethnic Festivals and National Festivals: Marginalising Minzu, Objectivising Han-ness .. 178 Conclusions ...... 188 Chapter 6: (Re)Performing the Inclusion/Exclusion of Uyghur-ness, (Re)Performing China .... 191 Introduction ...... 191 6.1: (Re)Performing Han-ness, (Re)Performing Inclusion / Exclusion ...... 193 6.1.1: Lineage and Self-identification Amongst Han in Ürümchi ...... 194 6.1.2: and Food as Boundaries ...... 200 6.1.3: Making Friends, Making Tuanjie...... 202 6.2: (Re)Performing Uyghur-ness, (Re)Performing Inclusion / Exclusion ...... 204 6.2.1: Discourses of Space and Identity Amongst in Ürümchi ...... 205 6.2.2: as the Centre of Uyghur-ness...... 207 6.2.3: Ana Til: Language and Uyghur Self-identification ...... 208 Conclusions ...... 213 Chapter 7: (Re)Securitising Ethnicity, (Re)Performing the Nation ...... 217

3

Introduction ...... 217 7.1: Securitising Han-ness, (Re)Performing China ...... 218 7.1.1: Securitising and Ethnicising Space Amongst Han in Ürümchi ...... 219 7.1.2: Han Ethnicisations of Violence in Ürümchi ...... 220 7.1.3: Saving the Nation: in Ürümchi ...... 224 7.2: Counter-Securitising Uyghur-ness, (Re)Performing China ...... 227 7.2.1: The Uyghur Masculinisation of Ethnicity ...... 227 7.2.2: Space and Security Amongst Uyghurs in Ürümchi ...... 229 7.2.3: Minzu Tuanjie and Security Amongst Uyghurs in Ürümchi...... 230 7.2.4: Language as Boundary Between Uyghurs in Ürümchi ...... 235 Conclusions ...... 240 Conclusion: (Re)Building the Nation ...... 243 Appendix 1: Lyrics to “One Family” Song – English Translation ...... 251 Appendix 2: Lyrics to “One Family” Song—Original Chinese ...... 253 BIBLIOGRAPHY ...... 255

Word Count: 109,979 Words

4

List of Figures

All photographs are taken by the author

Chapter 3 Figure 3.1: Photograph of image ““Bilingual Education” Hastens Buds (Shuangyu Cuibeilei)” from exhibition celebrating 60th anniversary of the People’s Republic of China organised by XUAR government……………………… 122

Chapter 4 Figure 4.1: Photograph of a erected by the city government’s planning department supervision centre stating “Minzu Tuanjie is prosperity, Minzu is disaster, taken in Ürümchi……………………… 147 Figure 4.2: Photograph of slogan “Strengthen Minzu Tuanjie, Strike Hard against ‘The Three Evils’ ”, taken on Tuanjie Street, Ürümchi…………………… 148 Figure 4.3: Photograph of a non-official slogan pasted around residential areas: “Xinjiang has since ancient times been an indivisible part of the motherland”, taken in Ürümchi……………………………………………………………… 151 Figure 4.4: Photograph of vandalised People’s Armed Police public display on “7-5” section “Regional Party Secretary Lequan makes his important speech regarding the ‘7-5’ incident”, taken outside Children’s Park in Ürümchi…………………………………………………………………………………… 160

Chapter 5

Figure 5.1: Photograph of MG car dealership advertisement with government slogan “The Han can never leave Shaoshu Minzu, Shaoshu Minzu can never leave the Han, and all Shaoshu Minzu can never leave each other”. Taken in a gents urinal in an Ürümchi coffee shop………. 169 Figure 5.2: Photograph of image “Alika Kedehan, ‘Loving Mother’ ” from exhibition celebrating 60th anniversary of the People’s Republic of China organised by XUAR government…………………………………………………………….. 173 Figure 5.3: Photograph of image “The mighty army – the troops of Xinjiang’s People’s Armed Police” exhibit from exhibition celebrating 60th anniversary of the People’s Republic of China organised by XUAR government…………………………………………………………………………………………. 176 Figure 5.4: Photograph of image “Ürümchi People’s Square” exhibit from exhibition celebrating 60th anniversary of the People’s Republic of China organised by XUAR government…………………………………………………………… 176 Figure 5.5: Photograph of patriotic stickers given out across the city of Ürümchi: “Love China (Zhonghua), Love Xinjiang”. 177 Figure 5.6: Photograph of two banners. Top slogan: “Joyfully celebrate National Day, happily welcome Mid-Autumn Festival”. Bottom slogan: “Strengthen Minzu Tuanjie, protect social stability”…………………………………. 181

5

List of Abbreviations

 CCP ()  ETIC (East Turkestan Center)  ETIM (East Turkestan Islamic Movement)  IR (International Relations theory)  PAP (People’s Armed Police)  PLA (People’s Liberation Army)  PRC (People’s Republic of China)  SCO ( Co-operation Organisation)  WDP (Western Development Project)  WUC ()  XUAR (Xinjiang Uyghur Autonomous Region)

6

The University of Manchester

David Tobin

Thesis submitted for degree of Doctor of Philosophy (PhD), 2013

ABSTRACT

Nation-Building and Ethnic Boundaries in China’s Northwest

This thesis will analyse the identity politics of the Chinese party-state’s nation-building project in the Xinjiang Uyghur Autonomous Region. It will examine how the party-state intends to overcome the barrier of ethnic boundaries in the production of a shared sense of multi-ethnic, national belonging. Uyghurs and Han can be thought of as belonging to different civilisations (Chinese and Turkic-Islamic) but in modern times they are often thought of as divided by ethnic boundaries. The party-state’s idea of the Chinese nation () is a nation-building project to eliminate these boundaries so as to produce a multi-ethnic nation. Fieldwork was conducted after the riots of July 2009 engulfed the region. Ethnically targeted violence against both Han and Uyghurs destabilised the city of Ürümchi and challenged the party-state’s vision of China as a unified and harmonious nation. The official Chinese explanation was that this was an internationally funded and synchronised terrorist attack but Uyghur rights groups have blamed tensions on government policy repressing Uyghur culture and stoking Han nationalism.

The theoretical framework employed draws from the concepts of production and performativity in Post-Structuralist and Critical International Relations (IR) theory, particularly the work of Cynthia Weber (1998) and David Campbell (1998). The critical approach adopted here takes security as a process of performative enactment of identity, which produces the units we take for granted as worthy of security. The analysis will examine official performances of what it means to be Chinese and Uyghur. It will then ask how these performances are received and (re)performed by members of the postulated nation. The party-state seeks to include Uyghurs as Chinese but it also excludes and securitises Uyghur Turkic and Islamic identities as ‘outside’ threats to the unity of the nation.

The research is a result of one year of fieldwork (September 2009-August 2010) in Ürümchi, the capital city of Xinjiang. This was the first ethnographic study of responses to the violence of July 2009. Furthermore, the incorporation of Han perspectives has been very limited thus far in the on Xinjiang. The analysis uses a top-down approach, which employs discourse analysis of official texts to understand what type of the party-state seeks to produce. However, these methods are coupled with a bottom-up analysis using ethnographic methods, particularly detailed, semi-structured interviews, to explore how these official discourses are received. The perspectives of Han and Uyghurs in Xinjiang can inform us how nation-building will unfold and what type of social dynamics it will engender. Analysing perspectives on the nation from below can help us understand the type of nation we expect to be produced in China rather than the type of nation the party-state narrates. The findings of this research demonstrate that both Uyghurs and Han are turning official Chinese nationalist discourses against themselves to articulate separate ethnic . Uyghurs frame China as an assimilationist transgression of ethnic boundaries for the benefit of the Han. Han frame their nation as under threat from Uyghurs and articulate China as a Han nation. The party-state’s nation-building project is unintentionally producing insecurity and reinforcing ethnic boundaries which remain obstacles to a shared sense of nationhood.

7

Declaration and Copyright

DECLARATION

No portion of the work referred to in the thesis has been submitted in support of an application for another degree of qualification of this or any other university or institute of learning.

Signed: David Tobin______

COPYRIGHT STATEMENT i. The author of this thesis (including any appendices and/or schedules to this thesis) owns certain copyright or related rights in it (the “Copyright”) and s/he has given The University of Manchester certain rights to use such Copyright, including for administrative purposes. ii.Copies of this thesis, either in full or in extracts and whether in hard or electronic copy, may be made only in accordance with the Copyright, Designs and Patents Act 1988 (as amended) and regulations issued under it or, where appropriate, in accordance with licensing agreements which the University has from time to time. This page must form part of any such copies made. iii. The ownership of certain Copyright, patents, designs, trade marks and other property (the “Intellectual Property”) and any reproductions of copyright works in the thesis, for example graphs and tables (“Reproductions”), which may be described in this thesis, may not be owned by the author and may be owned by third parties. Such Intellectual Property and Reproductions cannot and must not be made available for use without the prior written permission of the owner(s) of the relevant Intellectual Property and/or Reproductions. iv. Further information on the conditions under which disclosure, publication and commercialisation of this thesis, the Copyright and any Intellectual Property and/or Reproductions described in it may take place is available in the University IP Policy (see http://www.campus.manchester.ac.uk/medialibrary/policies/intellectual- property.pdf), in any relevant Thesis restriction declarations deposited in the University Library, The University Library’s Regulations (see http://www.manchester.ac.uk/library/aboutus/regulations) and in The University’s policy on presentation of Theses.

8

Acknowledgements

There are many people who have helped and supported me during the process of this PhD, too many to mention in this short space. My Masters, language year and three years of PhD was generously supported by the British Inter-University China Centre. I would like to thank that organisation for their financial support without which this PhD would not have been possible.

I am hugely indebted to my supervisors, Professor William A. Callahan and Dr Elena Barabantseva, for their patience and perseverance in guiding me through the production of this thesis. Their theoretical insights, writing suggestions, and professional advice have been invaluable. Bill allowed me the freedom to pursue a very difficult project in my own way. This would not have been possible under most supervisors, which I will never forget. Elena struck a difficult balance of being justly critical when necessary, while allowing me the room to develop my own avenues within an area in which she is more informed than I.

Enormous gratitude goes to Dr David Kerr and Dr Joanne Smith Finlay for invaluable advice and encouragement which without I may have never started this PhD. A huge debt of gratitude is also owed to Professor James A. Millward and Dr Gardner Bovingdon whose hard work has enabled a significantly easier life for a new generation of Xinjiang Studies scholars.

I would also like to thank staff members and colleagues at the Department of Politics at the University of Manchester in particular Stuart Shields, Peter Lawler, and Maja Zehfuss who have offered great support and assistance through these years. Special mention goes to the remarkably helpful and hard-working PG administrator Ann Cronley.

I wish to thank Tina McGeever, my mother, for her unfailing in me. Mum – your unfailing energy and fighting spirit never fails to inspire me.

To Tracey Fallon, as much as I should thank you for immense proof-reading and IT skills, it is your love which has often sustained me through these years.

For Ewan Mackenzie, your friendship and drumming have been a remarkable source of support.

Finally, and most importantly, to my friends in the years I spent in Xinjiang, it is you who have informed this project. You have challenged how I think about my own identity as much as the

9 politics of Xinjiang. Your courage in the face of adversity never ceases to inspire me. I hope the future holds genuine peace for you and that your knowledge can be shared with the world.

10

Dedication

For Michael Grey and Michael Tobin, whom without, this would never have been possible.

11

The Author

David Tobin is a lecturer in the Department of Politics at the University of Glasgow. David Tobin’s PhD research at the University of Manchester was funded by a British Inter-University China Centre (BICC) scholarship. David Tobin holds an MA with distinction in International Relations from the University of Sheffield and an MA with distinction in Contemporary China from the University of Manchester. He has undergone two years of training at and . Following a year-long fieldwork period in Xinjiang his article ‘Competing : Ethnic Unity and on China’s Northwest Frontier’ (2011) was published in the journal, Inner Asia. This reflects his general research interests: international relations, nationalism, ethnicity, and identity politics.

12

Introduction

The political scientist, Lucian Pye, famously conceptualised China as “a civilization pretending to be a state” (Pye, 1990, p.58). This idea built on the standard Sinological arguments of John King Fairbank (1968) that confrontation with Europe through the Opium Wars opened China’s doors to the world. In this , this confrontation left Chinese nation-builders facing the rupture of a Sino-centric East Asian inter-state system as China entered the Westphalian European family of nations. These ideas hold considerable currency in the People’s Republic of China (PRC) today. Zhang Weiwei, one of China’s leading political thinkers and former diplomat wrote in The China Daily, “China is unique in that it is a ‘civilizational-state’” (Zhang, 2011; 2012). Robert Cox (2000) framed civilisation as a temporary and contingent fit between constantly changing material conditions of political organisation and intersubjective meanings of , religion, and language. Following this understanding it is unlikely that any contemporary state, embedded in a world of state-centric security practices alongside global exchanges of people and goods, could ever be called a civilisation. Furthermore, in the Chinese case, civilisation in the singular is problematic because, as James Millward (2007) has shown, there are many groups within the contemporary borders of the PRC whose languages and are the result of thousands of years of trans-border, inter-civilisational exchange, which blur the lines between Chinese, Islamic, Turkic, Mongolian, Tibetan, and Indian civilisations. Perhaps China will become a civilisation in the singular. However, at the present Xinjiang offers us an example where a state offers its own model of civilisation (Chinese) to integrate a region whose peoples often think of themselves through alternative civilisational discourses (Turkic-Islamic).

Eric Hobsbawm (1990) claimed that China is an “historic nation” with such ethnic homogeneity that it has avoided problems of . This has been challenged by critical scholars of which show the diverse ethnic and civilisational groupings concealed by ‘Chinese’ history (Leibold, 2007, p.17-18). With China’s rise on the international stage, the party-state’s power to represent the multiple histories and peoples subsumed under the category of ‘China’ is growing. James Leibold tells us “it would seem that with each passing day, China is becoming more ‘Chinese’, and the of the frontier is fading into the red sun of the center” (Leibold, 2007, p.183). However, the mythomoteur, the which give meaning and structure to the nation, continue to articulate Xinjiang as a frontier when told from ’s perspective. Xinjiang is integrated into the Chinese nation but inscribed with difference and peripheralised to maintain its position as a frontier. This thesis seeks to problematise ‘China’ as a unit of analysis by exploring the persistence of the mythomoteur of Xinjiang, one of China’s frontiers. Its original contribution is to contribute to understanding Chinese nationalism by exploring how Xinjiang is articulated by the centre and how peoples in Xinjiang articulate themselves. At the level of international media, Xinjiang is not always deemed worthy of

13 mention in the discourses of China’s spectacular rise through reform and double-digit growth figures. However, the absence of explicit discussion should not be thought of as absence of social significance. The region’s name in Mandarin (Xinjiang) means “new frontier”. There is, of course, nothing new about it. Xinjiang was incorporated into the Qing state under the in 1759. However, this naming continues to articulate Xinjiang as a periphery, something new, against which the old centre of China continues to define itself.

This thesis will analyse the Chinese party-state’s nation-building project in the Xinjiang Uyghur Autonomous Region. It will assess how nation-building is received by peoples in Xinjiang. The time period of the fieldwork is highly significant. Fieldwork took place shortly after the riots of July 2009 engulfed the entire city as ethnically targeted violence against both Han and Uyghurs destabilised the region. According to official statistics, 197 people were killed (Xinhua, 2009c)1. Violence spread rapidly and chaotically across a city in which discontent with nation-building policies is widespread and different individuals with different grievances were involved. The violence of July 2009 became referred to as “7-5” in official documents, study-books, the media, and in everyday conversation. The official Chinese explanation was that this was an internationally funded and synchronised terrorist attack (Ministry of Information, 2009, p.3-5). However, Uyghur rights groups, including the World Uyghur Congress (WUC), blamed tensions on government policy stoking Han nationalism and creating an increasingly restricted public space for Uyghurs to express discontent (PBS, 2009). These were riots in which Uyghurs, Han, and the security services were all perpetrators and victims (Roberts, 2012, p.15-16 & 23). Grievances were held by Han as well as Uyghurs and individuals from both groups were victims of violence. The violence was inter-ethnic and included security services as much as ordinary residents. The region has long been studied for its “” between speaking ethnic Han and Turkic-speaking, Uyghur Muslims. However, this thesis will seek to de- centre this narrative and instead emphasise the multiplicity of perspectives which compete to be taken as the singular truth behind these events.

The thesis will examine how the party-state intends to overcome the barrier of ethnic boundaries in the production of a shared sense of multi-ethnic, national belonging. Understanding the party-state’s conceptualisation of the relationship between ethnicity and national identity will help us explore the boundaries of inclusion and exclusion in its model for nation-building. The success of this project can be gauged through how it is popularly received by individuals whose identity is being articulated through this model of Chinese nationhood. The analysis requires a top-down approach using discourse analysis of official texts and policy to understand what type of national identification the party-state seeks to produce. However, the

1 Chapter 4 will analyse these events in detail.

14 theoretical approach also requires a bottom-up analysis using ethnographic methods, particularly detailed, semi-structured interviews, to explore how these official discourses are received. A leading international scholar on ethnicity in China told me “Uyghurs can’t tell you anything about nation-building”. This is partly true because Uyghurs have little if any input into politics in Beijing. Every leader of the PRC and every party chief for Xinjiang has been a member of the Han . This helps us understand why ideas such as Fei Xiaotong’s (1988) “Han nucleus” which will develop ethnic minorities still hold so much weight in contemporary China despite ’s purported aim to eradicate “great Han ”. However, the very act of exclusion of Uyghur, and indeed all non-Han, perspectives from the centre of politics tells us that nation-building is exclusionary and non-voluntary. Furthermore, the perspectives of Uyghurs can inform us how nation-building will unfold and what type of social dynamics it will engender. Their peripheralised perspectives can tell us about the type of nation we expect to see in China rather than the type of nation the party-state tells the world we can expect to see.

Who is ‘China’ and who is this ‘we’, which ‘we’ want to secure?

During the 2012 Chinese leadership transition I was interviewed by the BBC several times for my thoughts on the new Chair of the Chinese Communist Party (CCP), . I prepared pages of notes on the impact of changes in social policy on , who after all will be the biggest winners and losers of any changes in the political environment in China. However, I was taken aback when I was repeatedly asked by journalists “should we be worried?” as if the ‘China threat’ somehow loomed large in Manchester, Coventry, and Cardiff. Xi Jinping’s careerism is such scholars do not expect any radical changes in China’s foreign policy2. However, the anxiety about what China’s rise means to our sense of we-ness looms large across Europe and the as the “Asian century” unfolds and power in international politics has long shifted eastwards3. The anxiety is as much as about shifts in power as it is about who we are. After spending three years living and conducting fieldwork across China my own sense of we- ness had been re-orientated so that I had to remind myself who was the “we” to which these journalists referred. The underlying nationalist ontology behind security discourses is such that “we” usually means other individuals who are ‘inside’ our national boundaries. In many ways this thesis tackles the idea of a China threat in a very different way from IR scholars4. It seeks to illuminate and problematise the power relations, which are concealed even in our everyday

2 For example, see Callahan, William A (2012b). 3 See Kerr, David (2011). 4 There is an abundance of literature on the China threat. Some of the more thoughtful contributions on these issues include Goldstein, Avery (2001), Johnston, Alastair I (1995; 2003), and Fravel, Taylor (2005; 2007).

15 thinking of who we are and who should be protected by security practices. It seeks to problematise state-centric assumptions about who we are and to whom we should extend our security concerns. As we will see, Xinjiang is subject to essentialisations from above which make alternative essentialisations the standard mode of resistance to the identity politics of the party-state. In A Treatise of Human Nature, David Hume (1739/2003) told us that “reason is the slave of the passions”. We scholars should embrace our normative commitments rather than wishing them away by omitting mention of them. The passion behind this project was to enquire why people seek to categorise and convert people in ways which run counter to and politicise those people’s self-identification. The analysis will use the multiplicity of essentialisations in Xinjiang to show that the party-state, and indeed no state, can monopolise the discourse of how people understand themselves.

It is nearly 30 years since Benedict Anderson’s Imagined Communities changed our theoretical landscape by suggesting nations should not be classified as but alongside social categories of religion and (Anderson, 1991, p.5-6). This thesis proceeds in this “anthropological spirit”. The taken-for-granted ontology of nations requires we take up ethnographic methods. Scholars of nationalism should listen to ‘ordinary’ members of the nation if we wish to understand the diversity concealed in these identity categories. Where the approach departs from Anderson is that is not seen as something out there. Instead, it shapes and is shaped by the nation, as well as religion and kinship. When I began this project, a leading Chinese studies scholar told me that this type of research in Xinjiang was impossible because “no one will speak to you”. They were correct that the politics of the party-state means that Uyghurs are anxious about open expressions of identity which could be construed as politically threatening to the CCP. However, throughout this research, I found that Uyghurs, and Han for that matter, are eager for the ‘outside’ world to hear their voices. The personal costs of speaking can be huge. It is these political restrictions, which contribute to the production of a social environment where two groups do not speak to one another. These ethnic boundaries continue to shape society in Xinjiang despite the party-state’s narrative that Han and Uyghurs are intimately bound together in 5,000 years of Chinese history. Time and time again interviewees would ask me what the Other said about their . The first time I was asked this I explained my research results because it stimulated dialogue. However, it quickly became apparent that this approach was unethical because I was exposing individuals to discrimination and stereotypes that the minimal dialogue between ethnic groups did not always reveal. Han and Uyghurs casually use so many stereotypes amongst themselves about the Other but the turbulent politics of the region are such that many of these narratives tend to stay in-group. My interviewees asked me because their interaction with the Other was ordinarily so minimal that they had no idea how they were viewed by the Other. Ethnic relations were already tense enough without my own interventions to add fuel to the fire.

16

Rather than think of nations as end-states, the approach here analyses the Chinese nation as an ongoing process of identity contestation. Nation-building is thus described as a project because it seeks to produce identities which as yet do not exist. Chinese nation-building is then a misnomer. It is not the nation-building which is Chinese. Rather it is the goal of nation- building for ‘China’ to become Chinese through regularised demarcations of boundaries and re- affirmations of the meaning of Chinese-ness. However, nations are not novel forms of consciousness and the radical juncture between pre-modern and modern forms of political identity are much exaggerated (Duara, 2009, p.97-102). ‘New’ political discourses take older identities and recreate them in new narratives of history (Duara, 2009, p.103). These discourses draw from repertoires that are neither old nor new but re-performances of older discourses in new forms which Kai-Wing Chow (2001) termed “semantic hybridity”. Signification is not a “founding act”. It is a regulated process of repetition where “agency” is located within the possibility of the variation on that repetition (Butler, 1999, p.185). As Karl Marx (1869/1963) wrote in the 18th Brumaire of Louis Bonaparte, “men make their own history but they do not make it as they please”. The thesis asks how people, not just men, in Xinjiang make their own history in the restrictive conditions which the party-state has established.

The framework applied here draws from the concepts of production and performativity in Post- Structuralist and Critical International Relations (IR) theory, particularly the work of Cynthia Weber (1998), David Campbell (1998), and RBJ Walker (1990). Performativity takes social categories, in our case security, identity, nation, and ethnicity not as natural givens but as discursive categories in a constant process of production. These categories are in constant states of becoming in processes of production through official articulations and popular reinterpretations of their meaning. Security, like identity, is not pre-given but is the ontological effects of practices which are performatively enacted (Weber, 1998, p.78). The critical approach taken here takes security as a process of performative enactment, which produce the units we take for granted as worthy of security. The approach follows the work of Judith Butler in conceptualising identity as unstable and socially contingent whereby boundaries are determined by processes of repetition that enforces its rules through the production of substantialising effects (Butler, 1999, p.185). (Re)performance is used here to conceptualise the response of the unofficial to the official. The suffix (re) is used to methodologically separate official party- state articulations of identity-security from unofficial Han and Uyghur responses. Using the concept of (re)performance enables the exploration of how official discourses are inverted against themselves by those they seek to identify. This thesis examines official performances of what it means to be a (Chinese) Uyghur. It then analyses the substantialising effects of these performances by asking how they are received and (re)performed by members of the postulated nation.

17

Power and Resistance in Textual Ethnography

The thesis will employ what Edward Said called a “textual attitude” (Said, 2003). “Texts” are everywhere and through references to other texts they articulate the meaning of and boundaries between Self and Other. The production of texts, a form of representation, is not simply an imitation of reality but “a practice through which things take on meaning and value” and through which ideas about reality are produced (Shapiro, 1988, p.xi). In the case of China in Xinjiang, the texts which will be analysed are found on the street, in the classroom, and in bathrooms. They are everywhere. They include policy documents, official news reports, textbooks, museum exhibitions, political , and indeed SMS texts sent by the party-state to all mobile phones in the region. The “omnipresence of power” helps us understand how power is exercised from “innumerable points” (Foucault, 1995, p.93-94). In Xinjiang the omnipresence of these texts reminds people of the power of the party-state to be everywhere. However, as we shall see these texts often unintentionally remind people that the party-state lacks the power to engineer people’s identities. The frequency of repetition would not be necessary were the party-state in a position to easily convince people who they are. These texts are not to be judged on the basis of their truth but on “the basis of their capacity for value creation in human beings” (Shapiro, 1988, p.11). This ontology understands human access to material reality as something which is mediated through the discursive structures within the texts available to them. Epistemologically, then, we learn about the social world through interpretive methods. To understand how reality is socially produced we have to interpret and get inside the discursive structures particular people read, how these are employed by power, and indeed how they employ power. The analysis here seeks to explore the relationship between power and resistance.

The concept of performativity has not been applied to identity politics in Xinjiang for a number of reasons. Many scholars, and most interviewees for that matter, dismiss the discourse of the party-state on Xinjiang as , which is out of touch with the reality on the ground. However, performativity as used here seeks to explore how official discourse shapes the reality on the ground. It assumes this shaping is complex and expects to find these discourses used for different ends which challenge the original intent of nation-building. Another reason the concept has made little inroads into studies of Xinjiang is that there are huge political costs at stake with regard to the fixed reality of identities in the region. As we shall see the party-state stakes much of its claims to legitimacy on the timeless reality of China and Chinese identity. One of the only tools Uyghurs have available to resist the project is to offer an alternative reality in which Uyghur-ness is as timeless and unbroken as Chinese-ness. The concept of

18 performativity does not think of identity as ‘not real’ but rather it re-frames identity as something which is articulated, maintained, altered, and produced through social practices, which are intimately bound into politics. Performativity is a to understand the productive effects of social practices instead of searching for their essential origins.

Analysing identity as a social practice helps us avoid the reification of the boundaries we seek to problematise. Viewing these social practices through identity politics means we have to consider the dynamics of contestation through power and resistance. Communities tend to respond assertively to encroachment on their boundaries because the boundary encapsulates the . They then feel under threat from an external source fearing that “if they do not speak out now, they will be silenced forever” (Cohen, 1985, p.108). The analytical aim is not to show how either the power of the party-state (nation-building) or resistance from below (ethnic boundaries) enjoys a superior ontological status over the other but rather to show how they shape and often reinforce each other. Power and resistance are not exterior to one another: “Where there is power, there is resistance” (Foucault, 1990, p.95). There is no absolute outside of power because a plurality of resistance usually takes place within the strategic field of power relations (Foucault, 1990, p.95-96). Power is inherently unstable because it relies on points of resistance to play the role of adversary to reinforce itself (Foucault, 1990, p.94-96).

James Scott’s (1985) Weapons of the Weak showed how legitimation gives marginalised groups the tools of resistance. Power and resistance are best understood through infrapolitics, defined as covert resistance which often takes place under the radar of social realms ordinarily thought of as political (Scott, 1985, p.336-338). Infrapolitics is of particular importance in authoritarian states where the costs of overt resistance are often so high that political alternatives are omitted from the official public record. The failure to take infrapolitics seriously in analyses of the former meant its collapse came as a surprise (Bovingdon, 2002, p.68). The grammar of legitimation is used as resistance in contemporary China against the party-state. The party-state’s claims to benevolence are countered with accusations of corruption and its claims to bring national glory to China are inverted by Chinese nationalists who claim the party allows the USA and to dominate world politics (Shue, 2004, p.28). However, infrapolitics is not used here to predict when or if the Chinese party-state will collapse. Ethnographic methods are employed to explore how top-down and bottom-up discourses of identity shape one another by imposing limits and constraints one another. These limits produce new meanings behind identity categories as resistance to official accounts take place. For example, the idea that Uyghurs are a Turkic nation is not new but it takes on new meaning when it is employed as a mode of resistance against the party-state’s discourse of 5,000 years of unbroken Chinese history.

19

Discourses on security and identity are part of the same process of articulating who we are and what threats are to be excluded from our self-identifications. Security seeks to secure our Self from the threat of the Other. Nation-building is characteristically framed by the party-state in terms of security because ethnic unity is framed as a means to a strong state and a secure nation. Nation-building is thus best understood as a means to define both the boundaries of who we are and where security must begin and end. Much of the Chinese Studies literature takes ideology and development as the key sources of the party-state’s legitimacy. The Othering of Japan and “the West” as security threats in the “patriotic education” campaign initiated after the Tiananmen massacre in 1989 is often taken as a key basis of the party-state’s legitimacy (Zhao, 1998; Callahan, 2006). Nationalism is a grammar which has come to be used as resistance to criticise the party for failing to protecting the interests of the nation (Gries, 2004, p.181). However, competing nationalisms in Xinjiang (official, Han, and Uyghur) complicate our accounts of China’s rise and its confrontation with external Others outside of the territory of the PRC. If we analyse the origins and practices of Chinese nationalism solely at the level of conflict with international Others (i.e. encroachment by ‘the West’ and war with Japan), we overlook the important role of ethnicity and internal Others in Chinese nationalism. Different nationalisms in Xinjiang articulate different Others and different selves.

“Domestic strangers”, or what Sun Yat-sen called “internal foreigners”, are important to understand the origins and the contemporary practices of nationalisms we label ‘Chinese’ (Callahan, 2010, p.128). Chow (2001), Leibold (2007; 2010), and Elliot (2011) have shown, the revolution to overthrow the Qing , which ruled ‘China’ from 1644-1911 strove to “expel the Manchu”, “kill the Manchu”, and “restore the Han”. This discourse placed ethnicity at the centre of nation-building. For revolutionaries, such as , the Manchu became the discursive Other thus excluding non-Han ethnic groups from the Chinese nation defined as the Han race (Chow, 2001, p.53). This conflict was then about internal and external boundaries at the same time. The Manchu were being excluded as foreign to the Han race and to Chinese civilisation but their and their imperial conquests remained part of the territorial unit of the ‘New China’. They were to become Chinese but could not become Han. The Manchu strangers, like other ‘barbarians’, became domesticated as internal Others. Thus, the boundaries of the Chinese nation cannot be understood solely through the securitisation of the external boundaries of the nation-state. The internal boundaries of the Chinese nation can be just as antagonistic as external boundaries between states. Neither the external (national) nor the internal (ethnic) boundary makes sense without the other. This thesis explores how inclusion and exclusion are indivisible strands of nation-building as a security problem.

Since China declared its own “war on terror” following the events of September 11th 2001 up until July 2009, Xinjiang had the lowest rates of violence and incidents of unrest than any other

20 part of China (Bovingdon, 2010, p.112). However, these statistics do not disturb the internal logic of representations of Xinjiang as the region continues to be popularly understood in China as hostile and dangerous. These representations help us understand that there is no objective relationship between danger and the event or mode of identity from which it said to derive (Campbell, 1998, p.2-3). The material ‘facts’ are that unrest declined, however, the idea that Xinjiang is a dangerous frontier is not one which has subsided. Despite, its relative stability compared to other parts of China, Xinjiang continues to be represented through danger and the “inside/outside Three Evils” of terrorism, separatism, and extremism. The “inside/outside Three Evils” denotes three ‘ideologies’ which the party-state deems undesirable. “The Three Evils” are said to threaten the territorial integrity of the PRC because they linking internal malcontents to external terrorist forces. However, much of the party-state’s discussion on the subject focuses on the level of identity. The “mistaken understandings” of “The Three Evils” are said to frame Xinjiang history as non-Chinese or Uyghur identity as Turkic and Islamic (Ministry of Information, 2009, p.47, 61, & 94)5. This discourse securitises how Uyghurs identify themselves because if they articulate themselves outside of (Turkic and/or Islamic) they are framed as security threats to the nation and to the state. Danger, thus, does not exist independently of those to whom it may become a threat (Campbell, 1998, p.1). This thesis is about identity-security- who does Xinjiang threaten and how is Chinese-ness being articulated through these securitisations?

Following the violence of July 2009, Alim Seytoff, spokesperson for the World Uyghur Congress (WUC) explained that tensions had been rising “because of the Chinese government’s political propaganda, of the Chinese people…and portraying Uyghurs…as terrorists, separatists, and Islamic radicals” (PBS, 2009). These claims illustrate some of the Uyghur concerns about the double-edged sword of Chinese nationalist discourses in Xinjiang, which seek unity but make those they claim to include feel excluded. However, these discourses neither emerge from nor have they been imposed on a tabula rasa. Instead they draw on and reconfigure older discourses of belonging in China. These discourses can be found at the ‘bottom’ of society amongst the marginalised as much as amongst elites. Instead of adopting the positivist epistemology of cause and effect, the analysis here seeks to explore how the official and unofficial mutually shape one another through the concept of (re)performance. (Re)performance is used to conceptualise the response of the unofficial to the official, which permits the examination of purported nations and not just nationalists at the elite level. This allows us to examine how these different identity discourses challenge each other by articulating alternative and often conflicting understandings of identity. Who reads the “texts” and how is of critical importance to understanding social life and political change. The enquiry

5 Chapter 3 will analyse this discourse fully.

21 thus does not simply ask what is the meaning of the text but asks who produces the text, who for, as well as who the text and how. We thus have to ask the nation and nationalists about nation-building. If we focus solely on the discourses produced by power there is a danger that we reproduce those boundaries which we seek to problematise in the first place. Instead, the approach here is to ask those who are officially articulated as members of the nation how they understand the ways in which they are identified by the state.

As Homi Bhabha tells us, the textual process of political antagonism, including the negotiation of the identifications offered by states, involves a “reading between the lines” (Bhabha, 2009, p.35). Once bounded and territorialised a nation may obtain what Bhabha called the “power to identify” but alternative self-identifications can turn texts back on themselves and challenge this power. The producer of the text can become the inverted, projected object of the argument turned against itself (Bhabha, 2009, p.35). Top-down approaches can help us understand what it is that people are “reading between the lines” before we ask how they do so. An ethnographic approach, using interviews and participant-observations, will ask how people are reading between the lines of official texts. This approach will help us to explore how official discourses on nationalism and ethnicity are being consumed, negotiated, or resisted. This thesis will explore how daily practices matter in politics and how politics is played out in these daily practices. Ethnography can be methodologically useful to the study of politics and the study of politics can be useful to conducting effective ethnography. What people say and do can be read as text, so the approach here is to explore the relationship between texts, usually those of power, and how people understand themselves in the case of Xinjiang often through modes of resistance to power.

Research Questions:

 What is the relationship between security and identity in China’s official discourses on Xinjiang?

 How is Xinjiang integrated into China and how do power relations shape the nature of this integration?

 Who does Xinjiang threaten and how is Chinese-ness being articulated through these securitisations?  What does the case of Xinjiang tell us about the relationship between ethnicity and nation in contemporary China?  How do official discourses on identity politics and unofficial self-identifications of Uyghurs and Han in Xinjiang shape one another?

22

 Is the securitisation of identity in Xinjiang producing a secure Chinese community? Or are ethnic boundaries being (re)produced in ways that contribute to insecurity?  What type of nation is China and how has it come to be thought of in this way?

Thesis Outline: Nation-Building and Ethnic Boundaries

This research aims to further our understanding of the multiple meanings of Chinese-ness and Uyghur-ness. It conceptualises the category of Uyghur as both a Chinese ethnic minority (shaoshu minzu) and as a self-identified Turkic ethnic group. The analysis explores the relationship between how people identify themselves (self-identification) and the ascriptive discourses of identity which define these people. This means taking seriously the discursive categories offered by the party-state from above. However, taking China as an unproblematised unit of analysis tends to reproduce ‘China’ as a fixed thing. In the case of Xinjiang, asking “what is this thing called ‘China’?” requires problematising this level of analysis. This is done here through enquiry at the level of the individual’s understanding of their own Uyghur-ness, thus far only researched in Uyghur Studies or social focussed on Xinjiang6. It is hoped that using theory from critical International Relations theory (IR) will help shed new light on Xinjiang’s ambivalent position in China. The hope is that discussions of Xinjiang can be brought into Chinese Studies and vice versa. This research will contribute to Chinese Studies by building on the expanding literature on Xinjiang to understand China. It will not analyse the region as an anomaly outside the Sinocentric theoretical frameworks drawn from what Uyghurs see as “Han China”. Instead, this thesis will focus on how the peripheral Other is central to defining the Chinese Self. There has been no published ethnographic fieldwork since the violence of July 2009. The unique contribution of this thesis is that it uses ethnographic methods to explore the relationship between official, Uyghur, and Han nationalisms in Xinjiang. To date there are no detailed ethnographic studies of Han in the region and this thesis will offer a way to explore how Uyghur and Han nationalisms both resist the party-state’s conceptualisation of China as a unified multi-ethnic nation. China is a narrative behind which competing meanings of Chinese- ness are contested, some of which seek to re-order the nation, some of which seek to articulate a need to depart from its boundaries and to escape from being forced to be Chinese.

Chapter 1 will review the theoretical literature in the areas of security studies and nationalism before exploring their applications to empirical cases in Chinese Studies and Xinjiang studies. This review will be used to position the theoretical and methodological approach adopted in the thesis in the broader literature of IR theory and Chinese Studies. Chapter 2 analyses the

6 For example, see: Bellér-Hann (2002; 2008); Dautcher (2009); Roberts (2004; 2009); Smith Finley (2000; 2002).

23 representations of the role of the ‘ethnic minority’ (shaoshu minzu) and Uyghur-ness as categories of persons within the official historical accounts of Xinjiang. This will explore how these representations in official texts articulate the boundaries and meaning of Minzu-ness and Chinese-ness. Chapter 3 will show how the party-state performs the domestic and the international in ways which make the internal and the external indivisible. Uyghurs are framed as a domestic threat because they are linked to cultures outside of China. Xinjiang is framed as an internal/external security dilemma for the Chinese party-state because it excludes and , important modes of Uyghur-ness, as un-Chinese and linked to an ‘outside’. Chapter 4 will explore how the boundaries articulated within the identity-security discourses of the party-state were performed, resisted, and (re)performed in the politics of the everyday following the violent events of July 2009. The argument presented in this chapter is that the meaning attributed to different violences against different ethnic groups is normatively ordered. These ordering practices articulated different levels of security threat dependent on ethnicity so as to produce a national model based on ethnic hierarchy. Chapter 5 explores public performances of inclusion of Uyghurs in China through the discourse of minzu tuanjie7. The argument presented in this chapter is that Chinese nation-building in Xinjiang is inherently exclusionary and hierarchically inclusionary. This paradoxical management of identity includes Uyghurs as part of China but excludes the ways they tend to identify themselves as un-Chinese. Through the example of minzu tuanjie, this chapter will show how nation-building in Xinjiang objectivises ethnic Han-ness as national Chinese-ness thus offering a model of conversion of difference rather than its inclusion. The thesis thus argues that nation-building is a model of conversion because it seeks to transform Uyghur identities, understood through Islam and Uyghur language, to become reframed through Chinese civilisation and Mandarin language. The omnipresence of these discourses reflects the party-state’s awareness that it must convert Uyghurs to China and persuade Han to include Xinjiang as part of Chinese-ness if China is to become a multi-ethnic nation-state.

Chapters 6 and 7 analyse the intersection of official and unofficial identity politics by exploring how the categories of Uyghur and Han continue to be (re)performed through daily practices. The in the final two chapters employs semi-structured detailed interviews with Han and Uyghur residents of Ürümchi to explore their perspectives on nation-building and their own Self/Other-identification. This is termed Self/Other-identification because it will show how both groups deploy discourses of Other-ness to articulate their own self-understandings.

7 This is officially translated as “ethnic unity”. However, following Harrell (1990) I have opted not to translate minzu as ethnicity because the concept does not entail self-identification. As we will see in chapter 5, translating tuanjie as ‘unity’ is also problematic. Translating tuanjie as unity conceals the ethnicised and hierarchical social relations it seeks to produce. This is not to say that nationalists round the world do not use discourses of unity to exclude alternative nationalisms. However, tuanjie has a specific ethnicised meaning which deserves analysis on its own terms.

24

Chapter 6 will explore how Han and Uyghurs in Ürümchi conceptualise the boundaries of community and belonging with reference to the official model of multi-ethnic belonging of Zhonghua Minzu and minzu tuanjie. This will examine the securitised boundaries of inclusion and exclusion, which Han and Uyghurs (re)perform in their self-selected categories of Self/Other-identification. Han employ pre-Communist discourses of lineage and ethnic Otherness through the division between Han and Inner China (neidi) on the one hand and Uyghurs and the “frontier” (bianjiang) on the other. Uyghurs draw from pre-1949 discourses of (Turk and Islam) to position their selves culturally outside Zhonghua Minzu through reference to daily, visceral experiences. The final chapter will explore how Han and Uyghurs conceptualise threats to their person and their sense of Self in ways which are shaped by official discourse but which resist this model of nation-building. The argument is that the party-state, Han, and Uyghurs all perform different identities through competing discourses of danger. Ethnic boundaries are being contested in terms of threat and security from above and below. The narrative of China is then imposed by the party-state onto this patchwork of competing self-identifications. However, nation-building fails to replace ethnic boundaries with multi-ethnic national belonging. The failure of nation-building in Xinjiang is partly because these boundaries have long become ossified and organise so much of public life. However, nation- building discourses of the party-state place the Han at the centre of nation-building so it fails to include different ethnic groups as equal members of national community. Nation-building, thus reinforces ethnic boundaries instead of dissolving them because Uyghurs feel marginalised through its inclusion and Han nationalists use these discourses to position themselves as superior to the ethnic Other.

(Re)securitisation and (re)performing nationhood are part of the same process of resistance to the party-state’s model of China. The key argument in this thesis is that the party-state’s securitisation practices exacerbate identity-security concerns amongst both Uyghurs and Han who see themselves as marginalised in their own nations. The thesis aims to problematise taken for granted identity categories. The history of Uyghurs and their social position in contemporary China show how China and are contingent, historically produced categories. IR should build a vocabulary to qualify “China” and “Chinese” so as to be clear what is “Chinese” about the phenomena we refer to. Furthermore, the ontologies concealed by this state-centric vocabulary are a normative problem. We reinforce the violently ascriptive identity categories given to us by states if we unconsciously adopt them to describe people who would prefer not to be included in them. To end this introduction on a cautionary note, the analysis here is about one China of many and one Xinjiang of many. It is nomothetic and idiothetic at the same time. The approach links macro-level analytical frameworks to micro-level fieldwork to explore the relationship between high politics at the national and international-level and the politics of the everyday in a particular space and time. Its weakness is that its conclusions are

25 not nomothetically applicable to all other spaces and times. However, such weakness is also its strength because the approach affords us a level of detail with which we can explore how politics affects people’s daily lives, which can remind us why politics matters. The argument that Chinese nation-building is failing in Xinjiang is shown by asking postulated members of the nation how they understand themselves. Ethnic boundaries are becoming hardened and popularly articulated as an alternative to the model of a multi-ethnic political community the party-state seeks to build.

26

Chapter 1: Securing the Nation in a Multi-Ethnic “Frontier”

The primary conceptual question in this thesis is how security and identity are inter-related discourses of nation-building. How do security practices articulate identities and how do discourses of identity articulate the need for security? This identity-security framework is located in Critical International Relations theory (IR) and will be used to understand the case of the contemporary Chinese party-state in Xinjiang and popular responses to its nation-building project. This analytical framework understands nation-building and security through the concept of performativity, developed by Cynthia Weber (1998), Judith Butler (1999), and David Campbell (1998). Nation-building is an ongoing social practice performing the boundaries of Self/Other rather than reflecting objectively existing ones. However, the thesis aims to explore how postulated members of the nation (re)perform these boundaries by turning nation-building discourses against themselves to articulate alternative understandings of nationhood. The party-state’s nation-building project in Xinjiang aims to transform self-identifications understood through ethnic boundaries (Han and Uyghur) into a multi-ethnic, self-identified national community (Zhonghua Minzu). The purpose of this chapter is to review the theoretical literature in the areas of security studies and nationalism before exploring its application to empirical cases in Chinese Studies and Xinjiang Studies. Secondly, this review will be used to outline the theoretical and methodological approach adopted in the thesis.

The chapter is divided into four interlinked sections. The first section focuses on security and identity in IR theory to explore how the literature has redefined (inter)national security as a means to articulate identity and demarcate the boundaries of Self/Other. This section will establish the identity-security framework used to analyse primary source materials. The second section reviews the literature from studies of nationalism and nation-building to argue that nationalism is best understood as a process of identity performance from above and below. This analysis will then examine how this literature can be applied to the cases of China and Xinjiang as national and regional units of analysis respectively. This section will show how the party- state’s nation-building discourses seek to impose particular articulations (Zhonghua Minzu) of national identification as matters of security. Where the second section is more focused on the top-down articulation of boundaries, the third uses literature on ethnicity and Uyghur studies to understand how nation-building is popularly received. This reviews the literature from Xinjiang Studies, largely from social anthropology, to examine how ethnic boundaries persist in the region. These boundaries often contradict the official Chinese nationalist discourse which defines China as a multi-ethnic . This section will problematise ethnicity by

27 understanding it as a discourse, which requires analysis both of the boundary and the content which it encloses. Ethnic boundaries like nations are not givens but constantly produced and reproduced often through banal, everyday social and discursive practices. The final section will outline the methodology employed throughout the thesis and how it will combine top-down and bottom-up approaches to identity politics to understand how official politics and the politics of the everyday mutually shape one another. The argument presented throughout the thesis is that the politics of identity requires an analytical framing which focuses on how the official and the unofficial interact and shape one another through the contestation of identity boundaries. The original contribution of this thesis will be to complement the Post-Structuralist frameworks drawn from IR with an ethnographic approach to security. The purpose is to examine how the self-identifications offered in official identity-security discourses from above are negotiated and resisted amongst postulated members of the nation below.

Section 1: Security and Identity in International Relations Theory

1.1: De-centring Security Studies

The post-Cold War flurry of security studies literature, such as the Copenhagen School exemplified by Barry Buzan (1991) and Ole Wæver (1995) and Human Security expounded by Ken Booth (2005) has widened the concept of security and the field of security studies. Both schools have shown how the traditional conceptualisation of security is state-centric. This was most paradigmatically expressed by Barry Buzan in People, States, and Fear (1991) who argued that realist accounts of security fail to account for the threat to security posed to people by states. A state-centric approach to security overlooks what are now vital areas of study such as terrorism, ‘civil’ wars, and ethnicised conflicts. In post-Cold War cases of insecurity, such as , the former Yugoslavia, and the Horn of Africa, the principal agents and victims have been people, not states (Kaldor, 2006; Kolodziej, 1992, p.422). Threats to human lives, such as war, poverty, and climate change are global in nature such that state-centric security discourses fail to adequately theorise or tackle them (Walker, 1990, p.3). The case of China in Xinjiang will be used in this thesis to analyse how intra-state security dilemmas can emerge when states secure some identity discourses over others. Identity-security includes and excludes different groups in different ways so as to create a hierarchical organisation of social positions in the politics of the everyday.

The analysis in this thesis is located at the level of the societal or identity sector whereby global processes of migration, regional integration, repression and ethno-religious separatism are being increasingly conceived as security threats to the survival of community, the nation, or a sense of ‘we-ness’ (Buzan et al, 1998, p.119-121). Ole Wæver (1995) showed that sectors such

28 as identity are “securitised” by states. Securitisation elevates areas of social life to a level of importance seemingly beyond the political. This represents identity as an existential matter and a universally desired state (Buzan et al, 1998, p.4). Where sovereignty and territorial integrity are the referent objects of state security, identity is the referent object of societal security (Wæver, 1995, p.60). This thesis is thus located in a body of literature which aims to broaden how we understand security by analysing social realms outside and alongside official politics. It will contribute to this literature through an analysis of how people frame the referents of their own security in ways which shape official politics.

The idea of threat to community is employed by people and states in an era of globalisation because “only rarely are state and societal boundaries coterminous” (Buzan et al, 1998, p.119). It is difficult if not impossible to identify a single nation-state in the world today, which stimulates ongoing tension between different groups who wish the state to represent their nation. State-centric analyses can obscure that human self-identifications are multiple and contextual but this multiplicity offers a broader reservoir of identity discourses for states and people to securitise. famously cited China as a homogenous nation where national identity and state borders conveniently coincide (Hobsbawm, 1990, p.66). However, China has 56 officially recognised ethnic groups alone, ranging from the majority Han to Turkic Muslims to Tibetan Buddhists. The argument here is that a unified national community (Zhonghua Minzu) is then best understood as a discursive performance of boundaries which are securitised by the party-state.

Securitisation is used in this thesis to refer to performances of identity through ‘official’ security discourses and ‘unofficial’ cultural discourses. These performances aim to produce the nation and nation-ness by making territorial and cultural boundaries congruent and representing challenges and alternatives to them as security threats (Campbell, 1998, p.61-70; Shapiro, 2004). While inscriptions of coherence and identification of threats to the self are driven by official discourses of government, they are equally located in “unofficial” sites of the community such as art, film, and literature (Campbell, 2003, p.57). Discourses on identity-security, which articulate particular visions of nationhood, appear in areas of social life often associated with popular culture, such as public rituals, displays, as well as written texts. These are often overlooked from traditional studies of security but will be used here to understand nation- building and the performance of boundaries.

The use of ‘identity’ here does not imply a fixed, given of impermeable boundaries. This follows the Post-Structuralist turn in Social anthropology where social practices do not simply reflect but constantly (re)produce how people define the boundaries of community between Self and Other (Bellér-Hann, 2008; Roberts, 2004). Similarly, the use of the securitisation concept

29 does not imply the securing of pre-existing boundaries. Securitisation is best understood as a performative enactment, which seeks to articulate boundaries and produce people’s identification with them. Securitisation is a process where the state does not exert its power solely though military and economic coercion but through a “productive cultural governance that further institutionalises the borders between the self and the Other, between patriotic citizens and foreign enemies” (Callahan, 2007, p.7). Cultural governance is a set of “historical practices of representation” where the struggle for the state’s identity is located in but never fully controlled by the state (Campbell, 2003, p.57). The case of China in Xinjiang and Xinjiang in China will show how the boundaries between Self and Other and between domestic and foreign are blurry. Securitisation can position people within the boundaries of the state as foreign and outside the boundaries of the ‘real’ nation. For example, during the Cold War, the “Soviet Threat” in US security discourse was represented in cultural as much as geopolitical terms (Campbell, 1998, p.25). The Communist “sympathiser” within represented Communism from the outside as an infection of the cultural coherence of the nation, thereby blurring internal and external security threats (Campbell, 1998, p.27). In our case the referent of official security discourses is the identity of ‘China’ as the multi-ethnic Chinese nation (Zhonghua Minzu). The integration of Uyghurs illustrates this blurriness because they are officially framed as what Callahan (2010) called “domestic strangers”. This thesis will show how Uyghurs are territorially integrated into China but are also represented through a need for cultural conversion so as to think of themselves and their history as belonging to China.

1.2: Performativity and the Production of Boundaries

To refer to Chinese security is to assume that the territorial are boundaries which can define who can be included in and excluded from such security concerns. Uyghurs in are consequently neither subject to nor protected by Chinese security in the same way as Uyghurs separated by only a few hours journey by rail in Xinjiang. Such security begins and ends with territorial boundaries. The ontology of this approach defines the referent of security and the boundaries between us and them by reference to state borders. Security discourses distinguish between the ‘self’ and ‘civilisation’ to be secured against the threats of the ‘other’ and ‘barbarism’ (Wæver, 1995, p.60; Callahan, 2004, p.27). Security then becomes a means of “telling us who we must be” (Walker, 1997, p.71-72). The analysis here then takes “who are we?” and “how are we to be secure?” as intimately linked questions which are best analysed together rather than leaving identity or security un-problematised.

The dominant discourses in the theory and practice of international relations are still tied to the concept of state sovereignty. However, state-centric approaches obscure the power relations

30 and the internal hierarchies, which shape how we define who we are. This state-centrism involves a spatial inclusion such that community and security are defined as inside states while difference and anarchy on the ‘outside’ define the relations between states and their peoples (Walker, 1990, p. 11-12). These categories of inside and outside have no essential meaning by themselves. Each side of this dichotomy is mutually constitutive of the other. Furthermore, they are stabilised by reference to other dichotomies such as self/other and security/anarchy. How Others are represented is crucial to how we represent our Selves and this shapes and is shaped by the security and foreign policy choices made by states. Security and the implicit Self/Other boundary, which it encapsulates is a discourse that “represents structures of authority and control” and its policies are based on representations of self/other through which “various forms of global otherness are created” (Shapiro, 1989, p.16-17). These representations of self/other and identity-security emerge in Xinjiang such that in China Uyghurs and the majority alike are to be bound together as a Chinese national community (inside) and secured from the threat of “separatism” (outside). Security does not reflect existing shared self- identifications between different ethnic groups but is a nation-building discourse, which aims to produce identity.

The critical approach taken here takes security as a process of performative enactments, which produce the units we take for granted as worthy of security. Security, like identity, is not a pre- given subject but emerges through practice. States and their boundaries are the ontological effects of practices which are performatively enacted (Weber, 1998, p.78). Identity-security is performed through sovereignty, neither of which have a pre-discursive existence. Security is best understood as a role where the “identity of the state is performatively constituted by the very expressions that are said to be its result” (Weber, 1998, p.90). David Campbell’s Writing Security (1998), like Cynthia Weber, drew from Judith Butler’s idea that the reproduction of gender takes place through the social practice of performing gender and is thus in a state of constant process (Campbell, 1998, p.9; Butler, 1999). Campbell shows how, through the demarcation of domestic and foreign, US foreign policy produces thinking on what constitutes self and other and what it means to be American. The self is determined by a regulated process of repetition that enforces its rules through the production of substantialising effects (Butler, 1999, p.185). In the same way that being defined as a male emerges from performing male- ness, being American emerges from doing American-ness. By articulating the boundaries between Self and Other, security discourses perform what it means to be a national citizen, in turn reproducing these boundaries. This thesis examines official performances of what it means to be a (Chinese) Uyghur as both a citizen of a particular state and as an ethnic group. Furthermore, it will examine these substantialising effects by asking how these performances are received and (re)performed amongst Uyghurs.

31

Security and danger, like Self/Other and inside/outside, are relationally defined and mutually constitutive dichotomies. Through “discourses of danger” states draw boundaries between safe insiders as the referents of security and dangerous outsiders as threats (Campbell, 1998, p.1-3). The Chinese party-state positions itself as the guarantee of identity-security against a contrasted past of international ‘humiliation’ and domestic disunity. This is exemplified in the official slogan “without the Chinese Communist Party, there would be no new China”. This discourse aims to reproduce the nation by demanding identification with the state as the guarantor against the threat of anarchy and danger from the outside (Campbell, 1998, p.12). Thongchai Winichakul’s Siam Mapped (1994) showed how exclusion and boundaries were equally as important as shared ideas or practices in the production of Thai national identity:

The creation of otherness, the enemy in particular, is necessary to justify the existing political and social control against rivals from without as well as from within. Without this discursive enemy, all the varieties of coercive force…would be redundant. In contrast to the general belief, the state and its security apparatus survive because of the enemy. (Winichakul, 1994, p.167)

The demarcation of group boundaries through the identification of “outsiders” and their association with danger are equally as important as shared ideas and practices amongst members of the group. Discourses on enemies and otherness, a “constitutive outside”, then produce the nation and promote social control to secure its boundaries. However, these ‘external’ Others are often political opponents who live within the physical boundaries of the nation. A key security strategy of the state is to identify those groups and characteristics which challenge or present alternatives to its conceptions of nationhood and to link them with external threats. This strategy allows the internal to appear external and national identity seemingly homogenous and unified (Winichakul, 1994, p.169-170). The analysis in this thesis will apply this framework to the ostensibly domestic issue of the articulation of nationhood through the official designation of security/danger and Self/Other within the boundaries of the Chinese state.

Despite the field of security studies’ increasing forays into identity and culture there has been little application of such theoretical developments to the case of the Chinese state in Xinjiang. Uyghurs within the boundaries of the Chinese state feel insecure because of the activities of the state such as the facilitation of Han in-migration, rapid development, and language policies (Becquelin, 2000; Smith, 2002; Dwyer, 2005). This thesis will explore how the state unintentionally creates insecurity at the ‘domestic’ or societal level of self-identification. This discussion will be continued in chapter 5 where the concept of securitisation is rethought as (in)securitisation. Its model of national inclusion excludes cultural characteristics such as Uyghur language and Islam as un-Chinese thus exacerbating fears of assimilation. Paul Roe (1999) shows how the ‘security dilemma’ in IR theory can be applied to intrastate ethnic

32 relations. As groups perceive other groups’ attempts to increase their own security as exclusive and threatening to their group this leads them to increase the security of their own group. This results in what will be called here a cycle of securitisation and counter-securitisation of group identification. In our case the primary groups involved are individuals who identify with the Uyghur and Han ethnic categories but also the state which claims they are all part of a singular multi-ethnic nation. Identity is securitised whereby the maintenance of a particular understanding of China’s boundaries is crucial to the CCP’s legitimacy as representative of the interests of a multi-ethnic nation. CCP policy is organised around its self-perceived frontiers and shaped by its confrontation with them (Callahan, 2004, p.27). The thesis thus aims to explore how identity and security are conceptualised through one another in China’s confrontation with Xinjiang, the “new frontier” and an un-Sinicised ‘outside’ but one whose conversion to China is framed as a security matter.

Section 2: Nationalism and Nation-Building

This section will examine how the literature on nationalism and nation-building can be utilised to show how official discourses on security articulate identity boundaries but are reconfigured from below. In our case these identity boundaries and the security they enclose are officially framed in terms of Chinese nationhood. Section 2.1 will show how literature on nationalism and nation-building can help us understand where security and identity intersect. These explore how group belonging is produced and under what conditions it is activated as a form of self-identification and a medium for action. Nations and nationalism rest on the reproduction of boundaries which divide humans into discrete political communities. These communities demand bounded identification with and the need to secure the Self before the Other. Discourses on Identity-security are thus framed as a form of nation-building because they articulate what it means to be and not to be a member of a particular national community. Sections 2.2 and 2.3 will review the literature on nation-building in China and Xinjiang respectively. This will outline the problematic nature of imposing the contingent category ‘China’ onto the entire history of the nation we call the People’s Republic of China today. For identities to be reproduced, people must internalise, share, and perform the officially articulated mode of belonging and self-understanding. For nation-building to be deemed effective, the meanings and self-identifications it offers would have to be shared by the subjects of the project. Nation- ness and the securitisation of its boundaries in texts have to be performed in the self- identifications and social expectations of ordinary people for it its goals of nationhood to be achieved.

33

2.1 Nationalisms and Nation-ness

Anthony Smith’s primordialist or “ethno-symbolic” approach defined nations as territorial communities of shared history and culture “founded on antecedent ethnic ties” (Smith, 1991; 1995). Primordialism has limited explanatory power to help us theorise the formation of modern, multi-ethnic nations, most notably the USA, , and China. Particular markers of ethnicity such as language may have their antecedents in China such as the existence of the semblance of from cave paintings circa 3,000 BCE. However, in the case of the Chinese state in Xinjiang, antecedent ethnic ties have limited applicability in any explanation of how the Han, an East Asian, largely atheist, Sino-Tibetan speaking ethnic group came to consider a region populated by Central Asian, Turkic speaking Muslims as an inalienable component of the Chinese nation. In the famous Warwick Debates, Ernest Gellner (1996) responded to Anthony Smith by showing that primordial, ethnic bonds may exist in some nations but their absence in others demonstrates they are ultimately inessential to the formation of nationhood. Membership of national community differs in kind from the fixed social classes of agrarian societies and presupposes the mastery of a literate, codified “high culture” above ethnicity and locality.

Benedict Anderson’s (1991) Imagined Communities extended the modernist, constructivist argument further to model nations as “imagined communities” which come to be “imagined” as “finite and sovereign” (Anderson, 1991, p.5-6). Nationalist literature, public , and museums imagined the nation by linking people who have never met into this bounded common community based on shared identification (Anderson, 1991, p.7, 25, & 36). National identity emerges from the socially constitutive process of imbuing “imagined communities” with the conviction that they are naturally and eternally linked, “stretching the short, tight skin of the nation over the gigantic body of the empire” (Anderson, 1991, p.86). However, nationhood is not an unambiguous social fact. It is contestable and continues to be a contested political claim in defining the boundaries of political communities (Brubaker, 1998, p.278). Nations, as categories, define political and social boundaries, which in turn provide the capacity for collective action as opposed to a guarantee of it (Brubaker, 2002, p.169). Who has the power to imagine or produce and securitise the boundaries of community and how they do so remain in constant contestation. In contemporary societies, this contestation has been accelerated rapidly through mass based literacy, which has enabled alternative articulations of nationhood to be disseminated and debated through print and electronic media. In one generation we have witnessed the growth of the supra-national European Union, the disintegration of Yugoslavia, the reunification of , the collapse of the Soviet Union, and the formation of Central

34

Asian states bearing the of their majority ethnic groups. It is thus more apt to frame political communities and their boundaries as in a state of on-going popular production through interaction between states and people rather than having reached an end state of being imagined by elites.

Nationalism is then often understood as the principle which seeks to make cultural and territorial boundaries congruent (Gellner, 2006, p.1). This idea drives the formation of nations, which subsequently emerge from this “high culture”. In this account nationalism is inherently state-seeking because it aims to redraw territorial borders to fit cultural boundaries. However, the conceptualisation of nationalism as state-seeking has less applicability to contemporary cases of nationalism such as the growth of far right political parties in Europe seeking to reverse migration or in China against the detention of Chinese fisherman by Japanese authorities in the Sea. Nationalism then as Rogers Brubaker tells us is not primarily state-seeking and should not be understood as arising from the demands of nations understood as bounded social entities (Brubaker, 1998, p.278). The demands of unrecognised nations may primarily focus upon statehood. However, for ethnic majorities, as we shall see amongst the Han in Xinjiang, nationalist demands are often directed towards internal boundaries and the meaning of the nation. Nationalisms articulate the meaning of the nation and seek to activate identification with these meanings. Group-ness, then, ought to be studied as an event rather than a referent of objectively real and bounded entities (Brubaker, 2002, p.168). In the case of Xinjiang, official nationalism does not seek to adjust territorial borders to correspond with cultural boundaries. Instead, nation-building seeks to transform culture so that people’s identity is congruent with the territorial borders of the PRC.

In the case of existing, modern, multi-ethnic states, the problem arises of who decides whose culture is to be officially deemed and continue to be deemed as “high culture”? What are the ramifications for pre-existing cultural practices and peoples who continue to participate in this community but are deemed to be low as opposed to high? Can people participate in more than one community? In the case of China, can the fifty six officially recognised ethnic groups be ethnic and national at the same time? Gellner’s focus on “high culture” and Anderson’s analysis of texts produced by elites overlook the significance of everyday politics and how ordinary people participate in the (re)production of national boundaries. In large scale industrial and post-industrial states many citizens will have little if any contact with elites and many opt not to participate in “high culture”. Michael Billig’s (1995) showed how nations are reproduced through daily “ideological habits” as much as grand scale performances of nationalism (Billig, 1995, p.1-6). The unobtrusive flag hanging unquestioned from a public building helps establish an “endemic condition” of nationalism and an idea of community which is taken for granted (Billig, 1995, p.6). National boundaries are often taken-for-granted and

35 naturalised ‘facts’, which define the inside and outside of community. However, the way these boundaries are performed by states and by their citizens (re)produce rather than reflect discourses of identity-security. If a sense of belonging is to be shared, reproduced, and successfully securitised within a community, it requires the participation and identification of its ordinary members.

An emphasis on the everyday does not mean we can jettison analysis of “high culture” or politics in explaining the reproduction of nations. It means we have to explore how official articulations of nationhood reach and are received by the wider populace. Individuals are embedded in multiple communities and re-imagine their own territorial and social boundaries, sometimes reinforcing official discourse and at other times challenging it. Vered Amit’s Realising Community (2002) critiqued the Andersonian shift in social sciences away from the study of community as constituted by social practice and toward imagined, conceptual communities. Community has always been something inherently “visceral”; pertaining to what and who we know and hold personal, emotional attachments towards (Amit, 2002a, p.1). For Amit, community is constructed through face-to-face encounters with members and outsiders which define and re-define its boundaries. Unlike imagined communities, in a visceral community, shared daily practices amongst people who do meet regularly are what produce a sense of belonging. Imagined communities are hugely significant but that what is imagined can only be felt if members are able to socially realise these imaginations (Amit, 2002a, p.8; 2002b, p.45). Nation-building, like power, is a relationship which requires analysis of its internal logic as much as its productive social effects.

Community and boundaries emerge in this ambiguous interface between the material and the ideational. The material and ideational are in many senses an illusory dichotomy (Doty, 2000, p.138). Material reality is never completely available to humans and our engagement with it is constructed through representational practices (Shapiro, 1988, p.11). Representation as a “practice through which things take on meaning and value” rather than simple reflections of reality encourages us to uncover the politics of the production of meaningful (Shapiro, 1988, p.xi & 11). Amit describes community as emerging at the interface between ascribed categories and face to face interactions and between imagined and visceral communities (Amit, 2002b, p.60). Nationhood, as a form of community, should not be confused with sociographic assertions of fact. It exists primarily in the minds of its members (Cohen, 1985, p.98). The boundaries of group-ness do not crystallise simply through the doing of social behaviour but in the thinking about what this behaviour means (Cohen, 1985, p.98). The reality of group-ness and identity-security are mediated by representation. The likelihood of the political activation of group-ness and the ability to draw upon its resources for collective action and identification are

36 mediated by prevailing representations. Nationhood is best understood not as a social structure but as a symbolic structure and as a form of popular representation.

If we take group-ness as a fluctuating social form then we see nationhood only provides the capacity for collective action rather than guaranteeing groups will spontaneously emerge from articulations of nationhood (Brubaker, 2002, p.167-168). Symbols, including nations, do not so much make meaning as provide the capacity to do so (Cohen, 1985, p.14). Communities tend to respond assertively to encroachment or transgression of the boundaries which encapsulate their identification because they feel their identity is under threat (Cohen, 1985, p.91). If the articulations of inside/outside, security/danger, and self/other run counter to pre-existing ones they can be expected to fail to be reproduced by the subjects of these discourses.

Meaning is a “personal, subjective, internal perception” and anthropological analysis should retain respect for individual cognitive processes (Rapport, 2002, p. 132 & 138). However, individuality and group-ness are often embedded in each other. Boundaries as the defining feature of we-ness which includes some and excludes others are defined by symbols. All symbols are “mediated by the idiosyncratic experience of the individual” (Cohen, 1985, p.14). Symbols or in Shapiro’s language, texts and representation, offer humans the capacity to make meaning and construct communities round the boundaries they offer. What is actually held in common between individual members of communities is not very substantial; “form rather than content” (Cohen, 1985, p.20).

The meaning behind boundary drawing symbols such as national flags, national day celebrations, or the Great Wall of China can vary widely amongst members. The possibility of flexible mediation of national symbols allows people to participate in and be integrated into the ‘same’ social practices without subordinating their individuality to orthodoxy (Cohen, 1985, p.20-21). Nationalism then can be an expression of group-ness and self-identity at the same time. Nationalism is most persuasive when it allows people to “refract their personal and local experience” such that “anyone’s can be substantially different from anyone else’s” (Cohen, 1996, p.807-810). Nationalism is least convincing if it alienates personal experiences by extending its domain to “people from whom one has always previously claimed significant difference” and impugning the integrity of their valued boundaries (Cohen, 1996, p.810). Following this argument, the securitisation of identity by states is unlikely to be effective in producing self-identifications if it transgresses existing social boundaries which are deemed to be meaningful by those who practice them. In our case, the production of Chinese nationhood is unlikely to succeed if it transgresses the boundaries of ethnicity between existing alternatives senses of group-ness between Han and Uyghurs. We can expect to see a heightened sense of self-identification and threat if this is the case.

37

The identity-security approach used here takes the role of the individual in any social analysis seriously. Individuals constantly interpret and reinterpret symbols via their idiosyncratic experiences and (re)produce the boundaries that give inside/outside and self/other meaning. However, this does not assume humans are the unencumbered selves of the Enlightenment. People select meanings from the representations available to them. Analysis of self- identifications requires the examination of the interface between text and social practice, “high culture” and the everyday, and group-ness and the individual. This thesis will explore how states “tell people who they must be”. However, it aims to go beyond state-centrism and ask how people see themselves and the interpellations which they face. The success or failure of nation-building is not simply dependent on the presence of performative boundary drawing. It is dependent on how people receive and re-perform these performances and (re)produce the symbolic boundaries of community.

2.2: Nation-Building in and of ‘China’

George Watson (2009) showed how popular literature on “failed states” conceptualises “nation- building” as something that developed, civilised nations should do to those nations which lack nationhood and specific structures of authority. Nation-building is about the imposition of order through the conversion of otherness and danger into something more familiar and secure. This vocabulary of nation-building is inherently colonial and mirrors the 19th century social Darwinist justifications of intervention through the dichotomy of civilisation versus barbarism (Watson, 2009, p.79). This teleological approach to imagines a linear form of towards capitalism and democracy. As we shall see in Chapter 2 the party-state similarly imagines cultural progress as a single teleological alternative but towards socialism and the internalisation of national self-identification.

Mark Berger (2006) critiqued the ahistorical analytical approach of the nation-building literature which assumes the nation-state and capitalism are parts of a “natural” global order. Chinese nation-building is equally problematic because it assumes the latent and natural existence of the nation it seeks to produce despite the competing narratives on history in Xinjiang. One way of incorporating competing historical narratives is through Charles Tilly’s (1992) work on the formation of the nation-state in Europe in Coercion, Capital, and European States. Tilly drew the analytical distinction between state-making as the social penetration of centralised, coercive control in taxation and security and nation-building as the creation of citizen identification with the centralised state (Tilly, 1992, p.1-3). This helps us avoid the assumption that nations have reached their end-state with establishment of political control.

38

Much of the literature on Chinese nation-building, however, rejects the theoretical models drawn from European experiences of nationalism. Prasenjit Duara’s (1988) influential work Culture, Power, and the State critiques Charles Tilly’s distinction between nation-building and state-making as inherently Euro-centric. Duara argues that unlike Europe, state-making and nation-building in China were simultaneous, inseparable historical processes. Early twentieth century nationalists such as and Zhang Binglin lamented the lack of a consistent name for the Sinic community. They linked nationhood, statehood, and modernity by arguing that a “race state” (guozu) was the most advanced stage of human evolution (Leibold, 2007, p.9-11). Following the anti-imperialist and anti-Manchu nationalist rebellions of the late nineteenth and early twentieth century, state-making in China was conceptualised within the framework of nationalism to unify the nation and modernisation to ensure its strength vis-à-vis European and Japanese (Duara, 1988; p.2-5). State-making and power on the one hand and moral regulation and legitimacy on the other are often co-constitutive forces. In China, literary production such as the writing of history and drawing of maps are the medium through which political power is exercised and the production of legitimate political meaning is controlled (Thornton, 2007, p.15, 203-204). Henrietta Harrison (2000) similarly showed that during the Republican era scramble for political control (state-making), the Guomindang (GMD) used public ceremonies, school textbooks, and the creation of a “sacred” to tell citizens how to dress and greet fellow citizens in a Republican nation-state. This drew the boundaries of and reconfigured what it meant to be modern and Chinese whereby the borders of the state were imagined as coterminous with the newly emerging Chinese nation (Zhonghua Minzu).

The scholarship of Duara (19888), Harrison (2000), and Thornton (2007) helps us understand how Chinese nation-building emerged. However, it is less successful in accounting for the enormous socio-political diversity, which characterise the histories that have contributed to producing what we call ‘China’ today. The territorial and social boundaries of what we call China today are quite different to what was called China in the 19th century let alone during the Han (206 BCE- 220 CE). We thus need different models for different regions of China, which have different experiences of Chinese rule and different social dynamics on the ground. Replacing Euro-centrism with Sino-centrism should not be considered intellectually or normatively progressive for it is merely another mode of essentialisation. Charles Tilly’s distinction between state-making and nation-building helps us avoid the problem of assuming the congruence of states and nations. This distinction has particular resonance for the case of

8 The later work of Duara (1995; 1998; 2009) is more focused on deconstructing the category of the nation in the study of history and is central to the framework developed in chapter 2.

39

Xinjiang. It permits examination of the conflicting national self-identifications of Han and Uyghurs without ignoring that the Chinese state has been made for some time yet continues to attempt to convince Han and Uyghurs that they are part of the same nation.

The pertinent question is not how did the Chinese nation build the Chinese state but how did so many diverse peoples come to be enveloped in this category we call ‘China’. The approach here then is to problematise the boundaries of the present by enquiring how the Chinese nation continues to come to be. This research proposes the ‘local’ case study as the most effective method to examine nation-building. Since the events of September 11th representations of “isolated, barren, and dangerous” regions have re-emerged under the banner of “nation- building” without reference to historical context or to the social dynamics prevailing in these regions (Watson, 2009, p.78). If we are to understand how states and nations are produced we must understand how they have been produced in the past (Berger, 2006, p.14). The specific historical and social backgrounds of each case are central to understanding how these projects will operate. If nations are self-identified communities then we must understand how postulated members of nations understand themselves. “Grand Theory” cannot account for particular socio-historical contexts nor can it predict the social dynamics engendered by particular cases of nation-building in particular regions.

The study of history in Chinese Studies has seen the celebration of what Paul Cohen (2010) called a “China-centred perspective”. This takes Chinese sources and perspectives as the starting point to understand Chinese history but does not address the problem of ambiguity with what we refer to when we say ‘China’. If we take the Chinese nation as our unit of analysis we are faced with the danger of taking the existing boundaries of the contemporary Chinese nation as a given rather than things which were and continue to be produced in a particular social contexts. The linear history of modern nation-states projects the category of the nation back in time as its actor which progresses from the past to the future (Duara, 1998, p.288). The category, ‘China’, conceals how different regions and localities within its boundaries are imagined in different ways and how these representations are regionally re-represented. However, to this day the Chinese state still has no universal orientation towards citizenship of different ethnic groups (Shih, 2002, p.236). The northern and of contemporary China (Xinjiang, , and Inner ) are still referred to as the ‘frontier’ (bianjiang) in official texts and popular usage today. In the Chinese imagination these regions were historically occupied by ‘barbarians’ (yi) but repackaged as Chinese ethnic minorities (shaoshu minzu 9) with the transition from empire to nation-state (Di Cosmo, 2002, p.2). There is a

9 Official translation of the term, minzu, has changed from ‘’ to ‘ethnicity’ (Barabantseva, 2009). This is in accordance with the political exigencies of the CCP’s shift in its legitimation from Leninist- Marxism to Technocratic-Nationalism. Following Stevan Harrell (1990), the term will be used here without

40 conceptual problem of applying the category ‘China’ to such regions, which are officially and popularly bracketed off from one another in stark discourses of Self/Other.

There is also a serious political problem that by taking ‘China’ as the latent nation to be revealed, this extends the domain of the nation to “people from whom one has always previously claimed significant difference” and impugns the integrity of their valued boundaries (Cohen, 1996, p.810). Focussing entirely on “inner China” (neidi), defined via its opposition to the ethnic Otherness of the “frontier” (bianjiang) can contribute to the conflation of the category China with the category Han and the objectivisation of Han-ness as Chinese-ness. A Sino-centric or state-centric approach overlooks the question of regional difference in the perceived legitimacy of the Chinese state. Regionally variable perspectives can help us understand where the boundaries of China lie as experienced and articulated by ‘Chinese’ people. Components of the so-called frontier deserve analysis on their own terms as their own centres. Furthermore, studying this ‘periphery’ can help us understand the ‘centre’ for neither centre nor periphery have any essence without the other, just as Inner China and the frontier can only be defined relationally.

The Chinese Studies literature helps us understand the centre from where nation-building emerges. However, it is less useful to help us understand the specific and locally variable social dynamics which emerge from these normative agendas of social control. The argument here is that if we approach the questions of the emergence or reproduction of China by taking the Chinese nation as our unit of analysis then our analysis is doomed to a Sino-centric teleology. The present would become a historically inevitable development instead of one of many alternatives. The present instead will continue to be contested through social alternatives while people articulate their self-identifications in different ways. The critical history literature on China’s contemporary north-west or the “Qing Frontier” (present day Xinjiang, Tibet, and Mongolia) seeks to ‘de-centre’ China by analysing regions according to their own historical specificities (For example see: Perdue (2005); Millward (1994; 2007); Leibold (2007); Jacobs (2007); Newby (2007), and Elliot (2001)). This regional approach to social histories within and across national boundaries rejects the ontology of nations as givens from the traditional Sinological approaches to ‘Chinese’ history.

The “Qing frontier” literature explores the regional variations of how the boundaries of nations are produced over time. This can help us understand how nation-building in Xinjiang is unfolding somewhat differently from Inner China. Such scholars employ a “Manchu-centred

translating it. This is because of the vast differences in meaning from European and American conceptions of ethnicity which imply both self-identification and civic membership in ways that minzu does not.

41 perspective” on the history of the “Qing Frontier”. This new approach aims to show how this empire-state departed from the vision of a Sinocentric cultural empire based on Chinese rituals and the “tribute” system exemplified in John Fairbank’s classic The Chinese World Order (1968). This literature shows how in the Manchu invasions of Mongolia, Zhungaria, Tibet, and the during the eighteenth and nineteenth centuries, these territories were understood and treated as military colonies long before they were re-imagined as part of any Chinese national community. This project of de-centring official historical narratives will be taken up more fully in chapter 2. It helps us ask ‘how’ questions associated with Critical and Post-Structuralist theory in IR: how has the territory of the PRC come to be called China? How is Xinjiang part of this nation? How do the boundaries of who and what is to be secured change with the boundaries of the nation?

James Leibold’s Reconfiguring Chinese Nationalism (2007) showed how the interaction with western during the 19th century may have triggered nationalist thinking in China as elites grappled with how to combat both the atrophy of empire and the threat of imperialism. However, there is a danger of overemphasising discontinuity between the Confucian moral ordering of traditional China and the geopolitical ordering of sovereign China today (Kerr, 2011, p.159). How competing conceptualisations of community from Europe and China became embedded in one another is best understood through Chow Kai-Wing’s (2001) idea of “semantic hybridity”. Semantic hybridity is a way of understanding cultural change by analysing how identity discourses become blended and create new meanings when they come into contact with new identity discourses rather than simply replacing one another as the singular mode of self-identification. Intercultural translation occurs in social contexts where ideological and political forces intersect to blend meanings and create new ones (Chow, 2001, p.48-50). The Han , Hanzu, is one such product of semantic hybridity because it emerged with intercultural translations of race and ethnicity from outside of China but became contested and imbued with meaning through the ideological struggle between Chinese revolutionaries and reformers in the early 20th century (Chow, 2001, p.48). Discourses of modernity in China represent a “reinscription of uniformity and difference”, which has generated resistance from frontiers both internal and international (Kerr, 2011, p.160, 167). Racial identities in the late imperial period were neither generated by a self-contained system called “Chinese culture” nor imposed through “Western” hegemony. They were created through cultural interaction with a variety of schools of thought by a group of reformers who actively responded to the decline of imperial cosmology (Dikötter, 1994, p.410-411). amongst Han Chinese elites was not simply substituted for modern nationalism when concerns of national sovereignty became commonplace for the first time during the late-Qing. Nor did Sinocentrism disappear when ideas of race were absorbed from imperialist European discourse. Nationalism and racism reinforced and reconfigured Sinocentrism at the same time. The conceptualisation of difference was

42 reconfigured such that the ideas about Sinic civilisation were replaced with the idea of the Han as a core, modernising group of a new “race state” (Leibold, 2007, p.9-11). Barbarians on the fluid frontier became ethnic minorities within the closed boundaries of the nation-state (Leibold, 2007, p.19-28). This transformation of peoples from outsiders to marginalised insiders blurred the distinction between China’s imperial domain and its sovereign territory. It included groups, formerly excluded and defined by their Otherness and danger to the Chinese state, into the inside of community and security producing new boundaries defining the Self/Other.

The highly influential early twentieth century thinker, Liang Qichao, built on the “Five constituencies” (Han, Manchu, Mongolian, Tibetan, Muslim) institutionalised under the Qing empire to recast them as races (Leibold, 2007, p.32). These races were active subjects of China to be assimilated into the superior Han core, at that time called Zhonghua Minzu (Leibold, 2007, p.32-33). Liang Qichao had wanted assimilation for the purpose of making the Han race stronger (Callahan, 2010, p.134). These ideas shaped the Republican Revolution and the overthrow of the Qing as visualised in the use of the “5 colour” flag of the Republic to represent the “5 races” (Harrison, 2000, p.101; Leibold, 2007, p.37). This discourse of the Chinese nation, places the Han as the centre of a hierarchy around which non-Han groups revolve.

The construction of a “multi-ethnic empire” (i.e. ‘China’) emerged from the tensions between the Han concept and the Qing territorial expansion in regions populated by other ethnic groups (Leibold, 2007, p.19). However, the institutionalisation of genealogy under the Qing became a primary marker of identity and a basis for racial thinking such as the Taiping Rebels labelling of the Manchus as a “race of demons” and the Emperor a “Tartar dog” of “barbarian origins” (Leibold, 2007, p.29). This insertion of race and the idea of a territorially bounded state reinforced Sinocentrism and gave leading thinkers such as Sun Yat-sen the vocabulary with which to argue that the greatest force of the nation is “common blood” and to label the Manchu a “different lineage” (yizu) (Dikötter, 1994, p.406; Leibold, 2007, p.30-31). Racial identities in the late imperial period were not generated by a self-contained ‘Chinese’ culture nor were they imposed through ‘western’ hegemony. They were constructed through interaction between many schools of thought who were responding to the decline of imperial cosmology (Dikötter, 1994, p.41-411). China’s resistance against imperialism then became a struggle between races and Yan Fu and Zhang Binglin defined the Chinese nation as a lineage descending from the mythical (Chow, 2001, p.53).

Han-centred models of China continue to be promoted today by the party-state and by Chinese social scientists. Fei Xiaotong (1988), China’s most famous social scientist, seminally argued that China has always been a pluralistic but singular nation, developed by the cultural and economic superiority of the Han “nucleus”. The history of Non-Han groups in ‘China’ is then

43 represented through assimilation described as “new blood” “for the Hans” (Fei, 1988, p.188). Ethnic minorities then are framed as a physical and social resource to be exploited for the sake of the strength and prosperity for the Han race. The concept of Zhonghua Minzu, as used by Fei Xiaotong and the party-state, no longer refers solely to the Han ethnic group but encompasses the 56 minzu who live within China’s contemporary borders. The Han then have been reframed from being China itself to being the “nucleus”. Zhonghua Minzu has been conceptually inflated to include non-Han groups as Chinese but in a way which maintains the economic and cultural superiority of the Han. This usage performs a ‘new’ understanding of what it means to be Chinese and conceals the socio-historical contingency of the meanings of these categories.

2.3: ‘Chinese’ Nation-Building in and of Xinjiang

There has been a tendency within Chinese Studies to overlook Xinjiang and simply assume it fits with research findings which take China as an unproblematised unit of analysis. For example, Rana Mitter’s (2004) highly influential Bitter Revolution is now used on most university courses on Chinese history and society. However, despite masterfully reviewing Chinese debates on the meaning of nationhood in the early 20th century, the book does not contain a single reference to Xinjiang or how debates on the role of Islam, the Soviet Union, and Pan-Turkism were contesting different conceptualisations of nationhood during this period 10. This exclusion of Xinjiang from analysis in Chinese Studies reflects and reinforces Xinjiang’s ambiguous position between sameness and difference as an integral part of China but culturally un-Chinese. There is a serious lack of expertise on the history and languages of the region. This is a particularly acute problem in the ongoing production of knowledge through the explosion of funding for Mandarin across the world in Confucius Institutes which reinforce the Sinocentrism of taking Mandarin to mean Chinese and Confucius as representative of the multiplicity of cultures within China. Publishing research on Xinjiang has also led to a number of scholars being banned by the party-state from returning to China, which creates professional anxieties regarding ongoing research access (Millward, 2011).

The separation of nation-building and state-making and the continuing political controls exercised over Xinjiang11 cast doubt on the popular understanding of “reform era” China as a distinct break from Marxist-Leninism on a path of linear progression towards open-ness and economic liberalisation. Rather than consider Xinjiang as an anomaly or an outlier, studying Xinjiang on its own terms can deepen our understanding of nationalism by integrating local and regional variations into how we conceptualise China. This regionalised focus does not exclude

10 For example, see: Brophy (2011) on the role of the Soviet Union, Schluessel (2009) on Turkish flags in Uyghur classrooms, and Millward (2007) on the Jaddids. 11 See Becquelin, Nicholas (2000; 2004a).

44 analysis of the national or the global. It helps avoid the reification of that which we seek to understand (‘China’) by challenging how scholars uncritically employ the category ‘China’ as a unit of analysis. Despite all the political and practical obstacles, there is an emerging and self- conscious field of Xinjiang Studies. This is perhaps most notable amongst historians and social scientists who focus on “multiple centres” and the “in-between-ness” of the region between China, the Soviet Union, and Central Asia (Perdue, 2005; Starr, 2004; Millward, 2007; Brophy, 2012).

Prior to the Qianlong Emperor’s conquest of Xinjiang in 1759 the value of the region was debated within the Qing court. Peter Perdue’s China Marches West (2005) illustrated how Many Han officials saw it as a barren wasteland, control of which was a drain on the Empire’s resources. Following military success, Xinjiang became seen as a source of glory. Xinjiang was only imagined to be part of China after military conquest and the establishment of political control. Xinjiang was once seen as a wasteful drain on state resources thus outside the realm of security and inside-ness before becoming Chinese. The politically contingent nature of the official Chinese representations of Xinjiang suggests that changes in the boundaries of what was to be called China altered who and what was to be secured. The Qing state did not seek to eradicate difference. Rather it re-inscribed and highlighted difference where Manchu elites sought to maintain separateness from the Han through policies, separate residences, and ethnic ‘nomenklatura’ in military institutions (Perdue, 2005, p.338). The Qing permitted the maintenance of Islamic law in southern Xinjiang and used co-optation of indigenous local leaders (begs) to maintain power (Perdue, 2005, p.339-342). The meaning of China changed with military expansion thus it makes little sense to stretch the category back to a time when its boundaries were unrecognisable from today. It thus helps to follow Tilly in theoretically separating state-making and nation-building when we write about Xinjiang.

It wasn’t until 1884 and the establishment of the “new frontier” province that Xinjiang became subject to Chinese civil law. After the conquest of Xinjiang, the Qing state continued to use different approaches of governance in different regions under the banner of “rule by custom”. This allowed the development of different languages, maintained existing local structures of governance, and institutionalized genealogical records as the primary marker of identification (Leibold, 2007, p.19-28). As Laura Newby (2005) and James Millward (2007) showed, the north and south of Xinjiang were never united as a singular political unit or seen as the same place in Chinese discourses until after the application of the realist military logic of Emperor Qianlong’s late 18th century conquest and naming of the region “new frontier”. After taking the North the Qing armies decided that taking the south would be necessary to hold the north. Laura Newby’s analysis of Qing documents indicates considerable inter-oasis trade, travel, and marriage across Southern Xinjiang (Newby, 2007, p.18). Culturally and linguistically similar groups from the

45 sections of Fergana Valley in present-day Uzbekistan, outside Xinjiang, were already referred to as “Andijanis” and outsiders (Newby, 2007, p. 24-25). Ethnic boundaries have a long history in the region. However, the Chinese integration of the north and south has produced a spatialised way of thinking of them together as a singular region.

During the Qing period, regional separation was then prevented by punishing disloyalty to the Emperor, which Leibold calls an “Emperor-Centric” system, as opposed to the Sino-centrism of earlier and Han Chinese officials, because it demanded loyalty to the state rather than identification with a group (Leibold, 2007, p.28). Under the Qianlong emperor, Tibetan and Mongolian religious leaders were co-opted and their religious symbols incorporated into the rituals of empire. However, attempts were not made to co-opt Islamic leaders into Imperial ritual or incorporate self-identifications in the northwest into China (Newby, 2004, p.42). The Qing thus created loyalty but not identity (Newby, 1999, p.459). The Islamic prohibition against nonbelievers acting as patrons prevented Muslims of Central Asia being fully incorporated into the Qing system (Leibold, 2007, p.28). This was state-making without nation-building and indicates that community was being produced in different ways in different places across ‘China’. As we shall see, this continues to resonate today through Xinjiang’s position of domestic Otherness in the Chinese nation. The role of Uyghurs in China is akin to what Sun Yat-sen called “internal foreigners” because they are Othered as distinct from and inferior to the Han Chinese majority.

At the time of the Opium War, Chinese nationalist discourses were emerging to simultaneously build a nation and to make a state against the imperialist ‘Western’ Other and the internal Manchu enemy12. However, at this time, the idea that Xinjiang was part of the Chinese nation which was being contested had not yet been articulated. In 1830 the Daoguang Emperor permitted Han Chinese settlement for the first time in Altishahr, the southern region of what is now known as Xinjiang. Inter-ethnic marriage continued to be outlawed and the Qing ruled via co-opted Begs (Clarke, 2007a, p.265). It is incomprehensible how a shared sense of belonging or community could exist while such laws separated different groups who could not even read the same languages. The Chinese state had already penetrated Xinjiang through centralised taxation and the military apparatus. However, the state had yet to incorporate Xinjiang into Chinese history and its peoples into the Chinese national community.

Literature from Xinjiang Studies and on the “Qing frontier” shows us that in Xinjiang, Chinese nation-building emerged after state-making. By 1947, the GMD made attempts to present Xinjiang as safe for Han residence in a series of imperialist visual representations of the ‘internal

12 For example, see Mitter (2004); Leibold (2007); Chow (2001).

46 other’ in the Tianshan pictorials (Tianshan huabao) commissioned by the Anhui-born General Zhang Zhizhong (Jacobs, 2008, p.553). General Zhang Zhizhong defected from the GMD to the CCP and then successfully negotiated the disbandment of the East Turkestan Republic in 1949 (Jacobs, 2008, p.548). For the first, a Chinese official claimed that Xinjiang was the home of both Han and Uyghurs (Kamalov, 2007, p.34). Zhang publicly criticised Han chauvinism and commissioned the Tianshan pictorials to visualise Xinjiang as an open, untapped frontier which welcomed Chinese migrants (Jacobs, 2008, p.551). Its people were represented unthreateningly as either elderly, bearded men or young, lively, naïve girls in colourful costumes welcoming the Han Chinese and the civilisation they bring with them (Jacobs, 2008, p.556). Xinjiang was being re-imagined as an integral and inside but peculiarly foreign part of the Chinese nation; a frontier on the periphery of the self. It was only with the collapse of Republican authority in the 1940s that General Zhang Zhizhong turned to historical appeals as a rationale for rule over Xinjiang. This was a first for the Chinese state. Zhang acknowledged that Uyghurs were indigenous to Xinjiang but claimed it was also the homeland of Han Chinese people (Kamalov, 2007, p.40). The boundaries of the Chinese national community were reconfigured in ways which indicate the contingency rather than any fixed essence of the nation. The nation was being performed so as to fix contingent cultural and social boundaries and map them onto the contemporary landscape of a world divided by sovereign, bounded states.

As Jacobs’ analysis of the Tianshan pictorials shows, there are long-standing tensions in official representations of Xinjiang between inclusion and sameness on the one hand and exclusion and difference on the other. Xinjiang today is represented as an inalienable part of the Chinese nation since ancient times yet it is still seen as a dangerous and barren ‘frontier’. This tension is captured in James Millward’s (1994) article on how the meaning of the “fragrant concubine” (Iparhan / Xiang fei) has been cast and recast according to the political exigencies of the time. The ‘fragrant concubine’ today has been re-appropriated by the state as a tale of “ethnic unity” (minzu tuanjie) 13 represented by her marriage as a Uyghur with a Chinese Emperor. The explanatory text at the tomb just outside Kashgar tells visitors that Iparhan’s entrance into the “imperial harem” expresses “the good wish for unity and mutual love between different since ancient times”14. This discourse not only reflects the forcible incorporation of Xinjiang into the Chinese state’s territorial jurisdiction but into the Chinese national community at the level of identity. Continued competition over the representation of the fragrant concubine as Chinese by the state and Turkic by Uyghur groups reflects how the expansion of the Qing’s physical boundaries of China have never been fully resolved in the realm of self- identification: “just as to Han minds, Xinjiang was in the empire but not yet of it” (Millward,

13 This is officially translated as “ethnic unity”. However, following Harrell (1990) we have opted not to translate minzu as ethnicity because it does not entail self-identification. As we will see in chapter 5, tuanjie is also a problematic concept. Translating it as unity conceals the political complexity of the hierarchical social relations it seeks to produce. 14 I visited the site three times in 2007.

47

1994, p.442). Iparhan is symbolic of minzu tuanjie and an integral part of the Chinese nation. However, she is still an ‘exotic’, ethnic Uyghur outsider. This is in the same way that Xinjiang, as a frontier, is included as part of the Chinese nation through the party-state’s discourse on minzu tuanjie but excluded from Inner China and the Han nucleus. Nation-building in Xinjiang today is a discourse which produces the notion of Zhonghua Minzu as a pluralistic, unified, multi-ethnic nation. This thesis will explore the contemporary tensions which remain between the inclusion of formerly-excluded “barbarian” groups on the “frontier” and the exclusion of those who understand themselves in alternative ways.

The party-state’s model of nation-building in Xinjiang is an identity-security discourse where articulations of Chinese-ness and security are performed through one another. The party-state’s discourse of minzu tuanjie, inclusive of all 56 minzu, performs the secure, inside of political community. Configurations of self-identification which challenge the idea of 56 secure, united minzu are represented as threats to national unity. Chapter 3 will show how the party-state conflates criminals, protestors, and poets through its discourse on “The Three Evils”. This is used in official media to explain the public burning of books on Uyghur history, clampdowns on campaigns for Uyghur language education, as well as the arrest of suspected terrorists (Becquelin, 2004b, p.45). Literary works and traditions which locate themselves out with the Chinese sphere or present negative representations of contemporary life in Xinjiang are hailed as security threats to the Chinese nation (Friederich, 2007, p.95). The boundaries of China are thus being securitised in attempts to restrict the productions of meaning attributed to the categories of nation and minzu. The securitisation of identity occurs in Xinjiang when the party- state claims that the promotion of Chinese language, protection of cultural heritage, and minzu tuanjie are critical to national security. Chapter 3 will show how those who do not participate in this model of modernisation and are then criminalised as threats to national security.

The securitisation of identity in Xinjiang cannot be understood by taking China as our sole unit of analysis. Within this discourse, the omnipresent threat to be excluded from the political community is a self/other producing discourse of danger in the representation of the backward, dangerous ethnic “terrorist” linked to Central Asia and Afghanistan. Chapter 3 will analyse how the Chinese party-state’s discourse on “The Three Evils” of “terrorism, separatism, and extremism” frequently uses the phrase “inside/outside” (jingneiwai) to link alternative self- identification to Central Asia and Afghanistan through discourses of danger. This is an attributive phrase to link disaffected Uyghurs who hold alternative, ethnicised conceptions of Uyghur self-identification with dangers from outside the legitimate political community. This brackets such self-identifications as external in a coercive mode of identity management, which demands identification with the Chinese state and its understanding of the Chinese nation as 56 united minzu (for example, see Bekri, 2008). The Chinese nation’s emergence at the confluence

48 of imperial and modern domains requires reading China’s territoriality from Beijing’s view and suppressing all alternatives (Sautman, 1997, p.364; Callahan, 2009). National identity, based on the pre-existing territorial boundaries of the PRC is officially represented as central to national stability and security. The “inside-outside Three Evils” (criminalisation) intensifies the apparent need for inclusive minzu tuanjie and the conversion of the Other to pacify this threat. This representation frames security and the self as inside Zhonghua Minzu. It is central to a nation- building discourse which seeks to produce self/other, inside/outside, and security/danger, in ways which cannot be applied to other regions of China. Xinjiang is performed and positioned between the self and the Other as a perpetual “frontier”. In the words of Gong Yufeng, spokesperson for the Chinese embassy in Egypt, Xinjiang is “a fairyland, ever and forever” (Gong, 2009).

Section 3: Ethnicising Boundaries from Above and Below

The project to build a nation, to tell Uyghurs who they must be, has been resisted frequently by both peaceful and violent challenges to the state’s authority throughout the 1990s (see Dillon, 2002; Bovingdon, 2003; Millward, 2007). As Michael Clarke (2007b) shows, the Chinese state has historically faced an “internal security dilemma” in the region. The goals of increasing security through further integration have always faced resistance as it has come into conflict with Uyghur identities thus decreasing “security” and destabilizing the nation-building programme (Clarke, 2007b, p.325). The final two chapters will analyse interview data to explore how members of the nation conceive security in ways which challenge the party-state’s securitisation of the boundaries of Zhonghua Minzu. Performing ethnic boundaries is the primary means for both Han and Uyghurs to resist the party-state’s model of ‘China’. This section will review the literature on ethnic boundaries in Xinjiang to establish and analyse the self-identifications the party-state seeks to convert through its nation-building project. The section is divided into two broad approaches to ethnicity in Xinjiang – top-down and bottom-up frameworks.

Researching ethnic boundaries in Xinjiang between Han and Uyghurs is a way of exploring how the identity-security discourses on minzu tuanjie and “The Three Evils” are popularly received in Xinjiang. These discourses aim to produce the secure inside of community through identification with a multi-ethnic Zhonghua Minzu, contrasted against the dangerous ‘outside’, defined through “The Three Evils”. Analysing ethnic boundaries allows us to examine how effective these securitisations of self-identifications and articulations of inside/outside are on the ground. Analysing ethnicity enables us to enquire if ethnic groups respond to attempts at integration by resisting this project and (re)performing ethnic boundaries instead of transgressing them. This

49 approach helps us understand how an internal security dilemma in the region unfolds in a process shaped by interaction between official politics from above and popular discourse from below. The purpose of using this framework is to ask if nation-building is successfully producing self-identifications such that ethnic boundaries in Xinjiang are being transcended. It also allows us to explore in what ways official discourses are challenged and how its securitisations are subject to counter-securitisations from below. The approach here does not essentialise ethnic boundaries as natural or given. It aims to explore how boundaries are performed and (re)performed through the interaction between states, identity categories, and social practices.

There has been a distinct and problematic lack of cross-fertilisation in the literature on identities in Xinjiang between two broad sets of literature which do not speak to one another (Bellér- Hann, 2007, p.2). The first is the top-down approaches of Sinological scholars and IR theorists. These scholars are often trained in Mandarin Chinese language and frame the region in state- centric terms of physical, territorial integration into China or majority-minority configurations as understood by the party-state 15 . The second broad set of literature is the micro-fieldwork approach adopted primarily by scholars in Central Asian Studies. These scholars are often anthropologists by training and schooled in Turkic and/or Russian languages16. These scholars understand the region in terms of social and cultural processes, which in the case of Xinjiang cut across the boundaries of nation-states. They frame the region and Uyghurs as Turkic, Uyghur, or Islamic and tend to be weary of the uncritical application of the official category of shaoshu minzu.

This division within scholarly approaches to ethnicity and nationalism in Xinjiang can be further elucidated through Yangbin Chen’s (2008) categorisation of studies of ethnic integration into the top-down “majority groups approach” and the bottom-up “minority groups approach”. The former focuses on how dominant groups set the goals of ethnic integration whereas the latter analyses the response of minority groups to ethnic integration. In studies of Xinjiang there are self-evident linguistic and methodological differences between these two types of literature. There are also epistemological and ontological reasons for their separation. The former consists of those mainly trained in political science understanding social change in terms of policy and physical integration with states as their unit of analysis. The latter understand social change and identities in terms of how people see themselves or “daily practices” (Roberts, 2004; Dautcher, 2009).

15 See: Blank (2003); Gladney (1990; 1996; 2004); Rudelson (1990); Schein (2000). 16 See: Bellér-Hann (2002; 2008); Dautcher (2009); Roberts (2004; 2009); Smith Finley (2000; 2002); Newby (2007).

50

It is important to note that in the work of many of the scholars discussed here that the top- down and bottom-up division tends to break down and should not be considered absolute. However, this division is useful theoretically to position this thesis in these debates but also because these divisions tend to act as starting points of analysis. Theoretical analysis requires a starting point. These starting points always exclude important dimensions of social life and scholars can only mitigate for such analytical absences because including everything at once is impossible. The narrative in this thesis begins with official discourse in chapter 2 but this is not where it ends. A synthesis of top-down and bottom-up approaches is necessary to capture the full complexity of the interaction between official and unofficial identity politics and between nationalists and the nation in Xinjiang. There is a growing interest in exploring the intersections between ascription and self-identification in Xinjiang 17. This thesis will synthesise these two broad approaches. It will explore how state and society, the official and unofficial, and competing discourses on nation-building and ethnic boundaries intersect to produce and (re)produce self-identifications.

3.1: Top-Down Approaches: Ethnicity as Ascription

Dru Gladney (1990; 2004) and Louisa Schein (2000) treat non-Han ethnic groups primarily as Chinese ethnic minorities. Their identities and political interests are understood by reference to their socioeconomic or discursive position within the Chinese state. Dru Gladney’s Dislocating China (2004) illustrates how the Chinese representation of minzu in both popular media and academic discourses as exotic, erotic, and primitive plays a pivotal role in the construction and essentialisation of the majority Han identity as both normal and civilised. These representations Other minorities as ‘backward’ and thus in need of the superior, modern, benevolent Han (Gladney, 2004, p.13-16). This projects an ‘external-ness’ upon non-Han as pitiable, dependent and to be absorbed into the civilised Han-sphere. However, these representations tell us more about majority identities than those of minorities. They have little to do with any real minority- ness but instead serve to produce a unified Self (Han-ness) against these representations of the Other (minorities). Louisa Schein claimed that “what constitutes the set of officially recognised nationalities in China today is the outcome of government policy during the last four decades” (Schein, 2000, p.88). Schein’s Minority Rules (2000) conceptualised official party-state discourses on the (Miaozu) of China’s southwest as “internal orientalism”. Schein shows how minorities are represented as agricultural, feminine, out of step with modernity but as integral components of Zhonghua Minzu (Schein, 2000, p.130). The category of Han is relationally represented as an economically advanced and civilising force. This discourse is exclusionary in its majority-minority binarism. However, it also represents a form of hierarchical

17 See: Bovingdon (2002; 2004; 2010) and Brophy (2011).

51 inclusion. The minority Other is forcibly included as a member of the “We” group (Zhonghua Minzu) but incorporated on a lower level of the social hierarchy (non-modern).

These representations of minorities can be understood through Edward Said’s understanding of the Orient as a ‘Western’ discourse and an “inferior self” to produce the ‘Western’ Self (Said, 2003, p.54). Minority-ness is a discourse of inferiority which reinforces representations of the Han majority as superior. These representations of the backward, ethnic Other are a form of nation-building. This form of nation-building is not simply about state-making or territorial demarcation but articulates the internal boundaries of the Chinese nation which hierarchically order ethnicity. This vision of social order articulates not only that people must be Chinese but what it means to be Chinese. The question emerging for this research is how these representations of external-ness from the modern core are uncomfortably integrated into an inclusive multi-ethnic nation-building project through which alternatives to this model of China are framed as security threats. Louisa Schein’s ‘internal orientalism’ helps us understand that in the ‘Chinese’ case this exclusion of the Other is also inclusive. These official discourses other shaoshu minzu as inferior. However, they also include them in Zhonghua Minzu and securitise these self-identifications by representing alternatives as threats.

Since 1955, the party-state has led an official project to categorise all China’s peoples into 56 ethnic groups (minzu). All of the party’s writings and official documents, including the constitution, describe an unshakeable de facto and de jure equality enjoyed by all ethnic groups in China (Constitution of the PRC, 2004). However, the party employs a binary division between Han (Hanzu) and “ethnic minorities” (shaoshu minzu) in all its literature and legal documents including the constitution and its white papers (for example: State Council, 2009a; 2009b). This fails to consider the enormous linguistic, religious, and ethnic differences between China’s “minority” groups ranging from (East Asian) Koreans (Chaoxiangzu) to (South East Asian) Thai (Daizu) to (South Asian) Tibetans (Zangzu) to (Inner Asian) Mongolians (Mengguzu) and (Central Asian) Uyghurs (Weiwuerzu). The category shaoshu minzu has no meaning without being relationally defined against the category of Han. Non-Han groups are defined “negatively” according to a lack of Han Chinese attributes in a hierarchical discourse of Chinese identity politics (Callahan, 2010, p.132). This dichotomising division is a form of Orientalism in which “there are Westerners and there are Orientals. The former dominate; the latter must be dominated”; a collective notion of identifying “us”, the Han as the “nucleus”, against “them” of the “frontier” (Said, 2003, p.6-7). The Self can only be defined relationally by excluding the Other and in a positional superiority to unify the Self and dominate the Other (Said, 2003, p.36). The Others are simply shaoshu minzu, with no reference to self-identifications or social practices, whereas the ‘Han’ self is granted a name. The discursive exclusion/inclusion of peoples formerly branded as barbarians into the Chinese nation is very much an imposition of

52 identity from the outside. This is what Gayatri Spivak (1988) termed “epistemic violence”’; an attempt for power to speak for the powerless. In our case we will explore how this epistemic violence includes and excludes Uyghurs at the same time.

Justin Rudelson’s Oasis Identities (1997) built on Dru Gladney’s (1990) “ethno-genesis” thesis and applied these top-down models of identity construction to Xinjiang. Gladney (1990) argued that the group-ness of Uyghurs was constructed by the Chinese state. The ethnonym Uyghur had not been used for five centuries since the conversion to Islam due to its association with the Uyghur Buddhist Kingdom in ninth to fourteenth century . The official re-invention of the ethnonym, Uyghur, then was said to impose a sense of “we-ness” onto disparate peoples. Gladney claims that the “” of the Uyghur was a “gradual evolution” through successive stages of integration into the Chinese nation-state, crystallising in 1934 with the adoption of the ethnonym Uyghur at a conference of the Xinjiang provincial government (Gladney, 2004, p.205-210). Rudelson (1997) then argued that despite the frequent anti-state and anti-Han incidents of violence during the 1990s, Uyghurs had no national identity. Rudelson claims that due to geographical obstacles, namely the Taklimakan desert and the mountains, Uyghurs traditionally had no inter-oasis contacts. They subsequently identified themselves purely by their place of birth. Rudelson’s fieldwork ended before the upsurge of violence and ethnic consciousness recorded by Joanne Smith in the 1990s, so his empirical claims required re-assessment (Smith, 2000; 2002). Perhaps more significantly, Rudelson’s theoretical framework employs an untenably singular view of identity. People understand themselves in multiple ways. Today many Uyghurs continue to speak with great pride of their birthplace and oasis-based stereotyping within the Uyghur community, both positive and negative, is commonplace (e.g. see Dautcher, 2009). However, local identities do not necessarily preclude ethnic, national, regional, or global self-identifications, which can co-exist in mutually informing and even contradictory ways.

Ildikó Bellér-Hann (2008) found that official native documents from Xinjiang 1880-1949 used the ethnonym, Turk, almost universally, which suggests a longer term shared sense of we-ness. Laura Newby (2007) shows that despite enormous geographical obstacles and contra Rudelson, inter-oasis trade and marriage actually flourished during the eighteenth and nineteenth centuries. Newby argues that group-ness preceded questions of national identity. As the most culturally and linguistically similar ethnic grouping to the modern day Uyghurs, were referred to as Andijanis and regarded as outsiders in the eighteenth century. Children from “mixed ” between Uyghurs and Andijanis were widely referred to as Chalgurts (‘half breeds’) suggesting ethnic consciousness prior to the existence of the Chinese party-state (Newby, 2007, p.18). Laura Newby (2007) effectively showed that “we-ness” does not necessarily require a fixed, state-sponsored ethnonym. These conclusions do not necessarily

53 lead us to abandon looking at the role of the state in the production of identification. However, we need to look beyond officially designated categories of identification if we are to examine how nation-building shapes and is shaped by how people understand themselves.

The approach to ethnicity used in this thesis follows Brubaker (1998) in arguing that group-ness as an event can form and disintegrate under different social conditions. This group-ness is felt amongst social groups and shaped from above but it is always in a state of process. Ethnogenesis and the political declaration of signifiers is only one dimension of this ongoing contested process. Ethnonyms merely denote categories, which encloses the vocabulary people within the category use to contest its meaning. They do not enclose meaning by themselves. Minzu are in some senses positionalities in which “minoritisation” may take place but it then takes a life of its own (Schein, 2000, p.96-99). Thomas Mullaney’s (2011) Coming to Terms with the Nation showed how China’s post-1949 ethnic classification project (minzu shibie) formulated an where the peoples China consists of 56 minzu and the party-state “scientifically” determines who belongs to what minzu. The decision was made to bypass subjective identifications in favour of “political exigencies” and “expert determinations” to make the undifferentiated mass of “alien peoples” within China’s boundaries intelligible to the state in a new way (Schein, 2000, p.82 & 95-96). The party-state’s ethnotaxonomy also ranked groups according the Morgan-Engels stages of development thus denigrating minorities as “less civilised” owing to their classification in earlier stages of cultural evolution (Schein, 2000, p.83). The widespread and unchallenged use of the official categories of Han and Uyghur tell us the state has a role at least in disseminating ethnonyms. This project has saturated all subsequent modes of Chinese ethnographic knowledge but its success depends on “unceasing efforts” by the party-state to intervene through education, museums, dance performances, and language policies (Mullaney, 2011, p.135). In other words, the party-state has relatively successfully engineered the existence of minzu categories. However, it has to engage with the constant processes of articulation, maintenance, negotiation, and resistance from below which shape the meaning of these categories in daily practice.

The importance of being Uyghur in Xinjiang has been transformed as Turkic speaking peoples have faced mass immigration of Mandarin Chinese-speaking peoples from other parts of China. According to official statistics the Han Chinese proportion of Xinjiang’s grew from less than 5% on the eve of the CCP’s rise to power in 1949 to an absolute majority in 2007, if military and Bingtuan 18 paramilitary farms are included (Rudelson, 1997; Xinjiang Uyghur

18 The Xinjiang Production and Construction Corps is popularly referred to as the bingtuan. It is frequently described by scholars as a state within a state. The bingtuan was the institution responsible for the communist-era “reclamation” of land in Xinjiang. The bingtuan’s unusual status as a quasi-state persists today, owning more than half of the arable land in Xinjiang and administering its own courts, police, schools, and hospitals (For example, see: McMillan (1979); Becquelin (2004)).

54

Autonomous Region Statistics Bureau, 2007). After 1949, Turkic-speakers were for the first time called Chinese instead of Muslims or barbarians. This demographic transformation coupled with the relatively recent project to convert the self-identification of Uyghurs has transformed an ethnonym into a security matter. Michael Clarke noted that there is a growing fear amongst Uyghurs that by being “swamped” by Han Chinese migrants their sense of Uyghur “we-ness” is under “threat” (Clarke, 2007a). It is not difference but the social organisation of difference which maintains ethnic boundaries (Barth, 1969, p.15). This thesis will present the argument that the social organisation of Xinjiang enables the free flow of people within the “modern nucleus” but excludes Uyghurs from this core thus producing China as an ethnic hierarchy. This hierarchy is concealed within discourses of minzu tuanjie but means Uyghurs experience Othering and exclusion while they are being included. What Stevan Harrell (1995) called the Chinese “civilising project” to convert minorities into China is complicated in Xinjiang by the exclusionary discourses of “The Three Evils”, which articulate the threat and possibility of inconvertibility presented by Xinjiang. Uyghurs respond to this exclusion by securitising Uyghur- ness as they define it.

The top-down frameworks of Gladney and Schein help us understand officially designated identities and the essential roles played by representations of ethnic minorities. However, their overwhelming reliance on Chinese sources means they tell us less about how different individuals within ethnic groups understand themselves and their roles in the Chinese nation. Understanding these self-identifications is central to helping us predict the success or failure of any nation-building project to convert people and produce meaning. As Stevan Harrell’s study of how the Yi of self-divide themselves into various ethnic groups shows, minzu are officially defined positionalities, which people may be born into but ethnic group-ness can only exist in action (Harrell, 1990, p.516). The group, the Other, and the state are three actors in negotiating the relationship between the category and the group. The nature of the group defines this relationship as much as the relationship defines the nature of the group (Harrell, 1990, p.516). Ethnicity and minzu are categories which conceal the ongoing contestation over the meaning and competition for power to articulate these discourses.

3.2: Bottom-Up Approaches: Ethnicity as Self-Identification

More recent scholarship on Xinjiang has forced scholars to rethink the state-led “ethnogenesis” argument. Uyghur national identity was officially articulated in response to the emergence of the nation-state in the region with the collapse of Russian and Chinese empires. In the 1910s and 1920s Uyghur in were central in articulating the category of Uyghur to identify sedentary Turkic speakers in Uyghur nationalist newspapers and in engagement with Soviet officials (Roberts, 2009, p.361). This culminated in the official Soviet

55 adoption of the ethnonym for sedentary Turkic speakers at the 1921 Tashkent conference to identify the newly acquired ‘minorities’ in (Kamalov, 2007, p.32). This was articulation from below to Soviet power above, which responded with the institutionalisation of the Uyghur category, which was then exported into the Chinese minzu shibie in the 1950s. Nevertheless, the debates between intellectuals and Kashgari traders in Soviet Central Asia had to proceed within the institutional bounds of the Soviet political system and the ideological framings of national revolution moving towards Socialism (Brophy, 2011, p.5). Uyghurs were involved in the articulation of this national identity in the Soviet Union from the outset, which contradicts the idea in the top-down approaches of Gladney and Schein that narratives of Uyghur identity construction should begin with the policy of the party-state.

The literature on identity politics in Xinjiang indicates not the disappearance of ethnic identities with economic development as the CCP’s historical materialism envisaged. Instead there has been a hardening of ethnic boundaries revolving around language, religion, and indigeneity as political, economic, and discursive inequalities persist and contribute to the maintenance of notions of separateness (Bellér-Hann, 2002; Bovingdon, 2002; Smith, 2002). The Xinjiang Studies literature loosely follows Chen’s (2008) minority groups approach by analysing Uyghur identities as distinct entities from the political and cultural nationalisms of the PRC. History and narratives told by Uyghurs are the focus of their analysis rather than Chinese ones. The debate within this approach focuses on whether the boundary or the social content which it encloses should be the focus of scholarly analysis.

The boundary-wave of this scholarship followed the outbreak of violence in the 1990s in Xinjiang when boundaries between Uyghurs and Han Chinese appeared fixed and immutable. Ildikó Bellér-Hann (2002) and Joanne Smith (2002) illustrated how during the 1990s Uyghurs in southern Xinjiang and Ürümchi respectively employed different strategies such as the Islamic pork taboo, Xinjiang time19, and dress codes, to reinforce and reproduce ethnic, social, and spatial boundaries vis-à-vis Han Chinese. Bellér-Hann (2002) and Smith (2002) use longitudinal perspectives to show how Han-Uyghur relations have become more antagonistic in the era of reform and modernisation due to the massive demographic transformation and subsequent employment squeeze and discrimination experienced by Uyghurs. These arguments all focused on Uyghurs as an ethnic group engaged in conflict with and or boundary drawing against Han Chinese as an ethnic group. The conclusions are that ‘China’ as a nation in Xinjiang is thoroughly divided by ethnic boundaries. Han see themselves as Chinese and Uyghurs see themselves as a distinct group outside this community. These conclusions perhaps reflect how the violence of the time reproduced and reinforced ethnic boundaries.

19 Xinjiang time is two hours behind Beijing time but is not recognised by the party-state. In their private lives, Uyghurs almost universally use Xinjiang time while all Han Chinese use Beijing time.

56

More recent content-based approaches have critiqued the tendency in this literature to view Uyghur self-identifications primarily as a binary construction of ethnic otherness vis-à-vis Han Chinese. Earlier boundary approaches did fail to capture the historical and contemporary complexities of multiple and contextual overlapping identifications in the region. More recent literature on Uyghur identity has seen an ontological shift away from the study of Uyghurs as an ethnic group and towards the study of social practices which constantly produce Uyghur- ness. Ildikó Bellér-Hann’s more recent Community Matters (2008) argued that the study of cultural boundaries was useless without the study of the cultural content which can define these boundaries in the first place. Jay Dautcher’s Down a Narrow Road (2009) argued that the focus on ethnic nationalism dominated the literature in the 1990s. This overlooked how everyday social practices help construct a shared sense of community revolving round face to face interactions in the home and in the neighbourhood. This new literature coincides with the return to studies of “visceral” communities in social anthropology over the conceptual, “imagined” communities mentioned earlier (Bellér-Hann, 2008, p.11).

Bellér-Hann (2008) and Dautcher (2009) highlight the need to study social practice in the production of community to avoid the reification of boundaries as intrinsically ethnic. These content-approaches help us understand the body of discourses people draw on to differentiate themselves in different social contexts. Scholars of boundaries ought to pay attention to these content-approaches in order to avoid reifying the boundaries they find in a particular time and place. However, the argument here is that neither the study of content nor boundaries alone are enough to grasp how people understand and change their understandings of themselves. Bellér-Hann and Dautcher perhaps over-state the case that the study of ethnicity and boundaries are no longer required to understand group-ness in Xinjiang. Without a boundary there can be no inside or outside and communities, ethnic or otherwise, would no longer exclude or include, they would dissolve into a human community. Boundaries do not have to be theorised as immutable and essential. They are socially contingent and contextually fluid. The meanings or content that form specific boundaries may change but boundaries themselves encapsulate groups (Cohen, 1985, p.91). The approach to the boundary here is informed by Frederick Barth’s Ethnic Groups and Boundaries (1969), which showed how ethnic groups employ self-ascribed cultural differences, in our case language and religion, as organisational vessels to construct and maintain ethnic boundaries. In socioeconomic and political systems which formalise ethnic identities and provide structures of interaction and unequal rewards based on them ethnic boundaries will be maintained (Barth, 1969, p.15). Nation-building aims at the collapse of ethnic boundaries but is organising ethnicity in a way which offers symbolic rewards of modernity and security to the majority at the expense of the status of minorities. Furthermore, ethnic boundaries are officially maintained across China through the application of

57 minzu categories in all public discourse including labelling on all citizens’ identity cards. The thesis then asks how these categories are negotiated and (re)articulated by the same people who are interpellated by them.

Section 4: Methodological Challenges From Above and Below

This section will outline the epistemological and methodological approach employed in this thesis which seeks to explore mutual shaping of identity-security politics from above and below. The section is divided in two parts to explain and theorise the two primary methods: discourse analysis of official texts on nation-building and interviews on self-identification with the postulated members of the nation. This section will also address and propose solutions to the problems faced in conducting research in Xinjiang. As we shall see, how we theorise these problems and the practical solutions we can employ in overcoming them differ greatly depending on whether identity politics are approached from above or below.

The methodology adopted in this thesis is constructed to explore how the seemingly objective world of official social categories intersects with the subjective and inter-subjective worlds of self-identification. This approach will be used to analyse how Uyghurs in Xinjiang receive official state discourses on Chinese-ness and Uyghur-ness. This adopts Edward Said’s “textual attitude”. Said took representations of the Other to be derived not from an empirically verifiable reality but from other texts which are written to categorise and dominate the Other (Said, 2003, p.92). In Gayatri Spivak’s terms such attempts by the powerful to speak for the powerless is a form of “epistemic violence”; a means to establish a narrative of reality as the singular and normative one (Spivak, 1988, p.76). Deconstructing the social categories of identification within texts designed to reproduce them can help us examine the political and social relations behind them.

To explore the self-identifications of those who do not have the power to disseminate texts, in the literary sense, requires observation and dialogue. A textual approach by itself can leave “the question of discourse’s address unaddressed” (Prakash, 1995, p.211). Texts and the identifications they seek to produce are best understood on the basis of their capacity to produce meaning. They establish a type of reality people often have to mediate in reproducing their own realities. Individuals live within a reality of ascription and Self/Othering but by negotiating and resisting the meaning of these categories these categories take on different meanings. Methods have to be adapted to fit our ontological and epistemological commitments. A social world constituted by interpretations requires an interpretive epistemology and methodology to study it (Hollis and Smith, 1991, p.407). For this reason the methodological

58 approach employed here uses ethnographic methods, particularly semi-structured interviews and participant observations. These explore how individuals (re)perform official discourse and how they identify themselves.

4.1: A Textual approach: Discourse and the Social Production of Reality

Researching the perspectives of the Chinese party-state and Chinese academics is comparatively straightforward as most are publicly available and politically promoted discourses. In terms of the performance of these official discourses, gauging which public displays to analyse was not particularly problematic because they are so widely advertised and well known. The purpose of this data collection is to examine how these discourses conceptualise the relationships between Uyghurs, Minzu, and Xinjiang on the one hand and the Chinese nation and Inner China (neidi) on the other. This will explore the role Uyghurs and Xinjiang are expected to play in the Chinese nation and help explore the tension between Xinjiang as both an eternal, integral part of the Chinese nation (inside) and a barren, often dangerous frontier land in need of opening up and development (outside).

The criteria for the selection of texts employed here are simply the most widely read texts in Xinjiang published by the party-state. The texts were not selected for their truth value or academic rigour. They are selected because they are powerful official representations which people have little choice but to engage with. This selection reflects the focus of this thesis on the intersections between ascription and self-articulation. Political education (zhengzhi jiaoyu) in Xinjiang is so extensive that every official body from the police to universities comes to a complete standstill every Wednesday afternoon. They are closed with signs posted on doors such as the Public Security Bureau announcing “Closed this afternoon for political education”. This presents constraints as well as opportunities for researchers. After failing to gain access to official archives, one can access these mainly publicly available materials studied by all employees of the state, cadres, teachers, and students. These resources are rarely systematically examined by scholars of Xinjiang because they are easy to dismiss as “propaganda” with no truth value. However, they are an invaluable means to explore the party- state’s nation-building project and how it seeks to produce particular forms of self-identification.

Many of the key conceptual arguments explored in the thesis such as those of “The Three Evils” and minzu tuanjie discourses are saturated in public discourse through television, newspapers, and public speeches. Footnotes will be utilised to present additional sources for these discourses so as to show they are not limited to an anomalous or unusual texts. Particular use

59 will be made of resources from the party- state’s region-wide drive for minzu tuanjie education (minzu tuanjie jiaoyu) following the riots of July 2009. Through interviews and discussions with teachers and students of political education, the core texts within these courses were identified and selected. These materials are used to “promote ethnic unity consciousness … with patriotism as its core content” (XUAR Government, 2009, article 3). These classes are compulsory in all schools, universities, public bodies, and for all employees of the state. Most of the materials are used in lectures and training at the level of the work unit (danwei). This means that almost all residents of Xinjiang will have some familiarity with these discourses even if they chose not to consume public media. These texts include those produced at the national level for a national and often international audience, particularly “White Papers” such as the State Council’s China’s Ethnic Minority Policy and the Common Prosperity of all Ethnic Groups (2009a) and The Progress and Development of Xinjiang (2009b) (hereafter: Ethnic Minority Policy and Progress and Development).

The second type of text is those produced specifically for consumption in Xinjiang through minzu tuanjie education. These are often published by the State Council or the Ethnic Affairs Commission in conjunction with the Autonomous Regional government. One such text utilised here is The Xinjiang Cultural Knowledge Study Guide (Hereafter: Xinjiang Cultural Knowledge). It is a text specifically devoted to training local cadres to understand the history and cultures of Xinjiang. However, it is also used in minzu tuanjie classes for other social groupings and was publicly available for a meagre 12 renminbi under the ‘hot topics’ section of Xinhua bookstore in the centre of Ürümchi.

Within this category of Xinjiang-specific texts, there are two texts which are not available for public purchase which will also be utilised. The first is The 50 Whys: Protecting National Unification, Opposing Ethnic Separatism, Strengthening Ethnic Unity Study Book (hereafter: The 50 Whys). During the fieldwork period (2009-2010), this text was published by the Ministry of Information and was studied by all teachers and students at every higher educational institute in the region. Many informants explained to me how they had to attend lectures reciting its content and pass exams based on memorisation of the key arguments in order to continue their education. The second text, Common Knowledge of Ethnic Theory (hereafter: Common Knowledge), is used for all middle school children aged 12-15 in Ürümchi as part of Ethnic Unity Education. It was taught and assessed in the same style. These two texts are marked for ‘internal circulation only’ and had to be acquired through making connections with staff at various institutions20. There are certainly no intelligence or military secrets of any value in these

20 Originally, The 50 Whys came into my possession by chance when I asked a senior teacher at a university if I could borrow the copy on his desk. He seemed eager to express how interesting it was and that I would learn about China from it. However, I was subsequently chased on campus by a Communist Party University supervisor demanding I return the book. I was subsequently questioned on what I knew

60 texts but they do offer a more systematic treatment of the issues of minzu and national identity, which is why they were particularly useful for this research. They also make certain claims, which would likely be surprising for people from other parts of China and alarming to sections of an international audience, particularly religious groups and human rights NGOs. As we shall see, these include claims such as the need to promote religious freedom by banning students from attending , framing opposition to Mandarin Chinese-medium education as terrorism, and stating that ethnic identity must be assisted to fade away through “ethnic extinction” (minzu xiaowang).

4.2: An Ethnographic Approach: Addressing the Question of Discourse’s Address

Unlike more traditional anthropological studies such as Jay Dautcher’s (2009) Down a Narrow Road, this research does not aim to capture the totality of social experience of Uyghurs or their community. It aims to understand a particular aspect of Uyghur identity. The focus is Uyghur political identity - how they understand China and how China understands them. This study is of a time and a place, specifically post-“7-5” Ürümchi with some earlier references for comparison. This research is about identity narratives and narratives about identity as articulated by and for Uyghurs. The history of Ürümchi can be understood as a “frontier” city (Ethnic Affairs Commission, 2009). Most of it was built by Manchu settlers and officials unlike the famous Uyghur-majority Oases of Kashgar, Khotan, and which enjoyed if not independence until 1949. It then occupies a position of in-between-ness. Uyghurs do not see it as “their” place. Many deride it as “Hanified” and dirty with no “real” Uyghurs. Han migrants from all over the country converge here looking for work.

During a year of fieldwork (September 2009-August 2010), I engaged with local residents in Xinjiang and consulted online blogs and forums to determine which are the most popular and well regarded unofficial forms of popular culture. Given the political sensitivities in the region, conducting interviews is considerably more challenging. Despite China’s post-Mao ‘Opening to the World’ and the accompanying re-opening to foreign social scientists, the conditions for conducting research in the PRC are increasingly tied to locality (Smith, 2006; Thogersen and Heimer, 2006). There are many political obstacles to conducting social research on Uyghur identity in Xinjiang which reflect the unevenness of China’s modernisation project and ‘opening to the world’. These obstacles inevitably shape any research project in the region. Like any research project compromises have to be made between ideal methodological rules and the about the book and why I wanted it. The following day, another teacher laughed the incident off and told me that they didn’t want it back because it was “secret”. It was just that it is “only for Chinese people, foreigners wouldn’t understand it”. Later, another staff member was willing to give me a copy because they wanted me to “know the truth about 7-5” and “what the Uyghurs have done to us”.

61 reality of restrictions in the field (Thogersen and Heimer, 2006, p.3-4). Solinger (2006) and Thogersen and Heimer (2006) argued that unprecedented access is now open to foreign social scientists across China and that Chinese people are more willing than ever to discuss matters of culture and politics. However, the primary problem in conducting research in Xinjiang is access due to the politicisation and securitisation of identity in the region. Formal interviewing on sensitive identity topics in Xinjiang remains an unrealistic option (Smith, 2006, p.135). However, this project is not methodology-driven social science. The research questions and their ontological commitments shaped the methodology rather than the other way round. Therefore, it uses the methods available, which help answer the types of ontological questions which problematise categories of identity.

Gaining access to getihu21 and academics usually involved offering the interviewee benefits. In the case of the latter this was usually English language tuition and for the former, the purchase of goods or services. Herbert Yee (2003) is perhaps the only foreign scholar since the 1990s to have used official government channels to collect data on ethnic relations in Xinjiang. However, his work was marred by this route as he had to drop many original questions. In three weeks of fieldwork, his ‘random’ sampling provided by official channels found seven “mixed race” marriages. This can be compared to the 20 months of fieldwork conducted by Jennifer Taynen (2006), which failed to encounter a single mixed marriage. This illustrates how being chaperoned by officials to examples of a ‘’ will lead to spurious conclusions regarding that society. This type of research is further limited because the room for qualitative methods is nil. A discussion beyond the agreed, set questions to gain understanding of why people see things the way they do would not be permissible (Gold, 1989; Smith, 2006). In Xinjiang Studies Taynen (2006), Smith (2000 & 2002), Bellér-Hann (2000), and Bovingdon (2002; 2010) follow Thomas Gold’s (1989) “guerrilla interviewing” technique by employing “informal discussions” and participant-observations in their data collection which has the advantage of being free from state interference. “Guerrilla interviewing” is defined as “unchaperoned, spontaneous but structured participant observation and interviews as opportunities present themselves” (Gold, 1989, p.180). This research uses ‘conversations’ in ordinary settings after having built enough trust to gain answers not possible in a more formally politically regulated social context. This requires building sustained relationships but also the ability to empathise and express understanding with informants for them to “open up” (Burgess, 1982; Gold, 1989; Smith, 2006; Solinger, 2006). These ethnographic methods have the advantage of incorporating ‘everyday’ life into the conduct of field research (Burgess, 1982, p.107; Gold, 1989, p.182). These ordinary, daily settings allowed me to avoid more formal

21 Getihu refers to self-employed entrepreneurs and here is used to refer to small-scale traders or the self- employed who employ family members or up to one non-family member.

62 settings which can lead to standardised, formalised answers and in this case would be influenced by the distrust that the researcher is in fact aligned with the state.

Methodologies employing such ‘conversations’ often appear to be without structure to the interviewee but through thorough preparation and background knowledge the researcher establishes a framework within which the ‘conversation’ takes place such that it is flexible but also controlled (Burgess, 1982, p.107). With set questions and concepts in mind, the researcher then proceeds in a “progressively structured” manner (Smith, 2006, p.136-137.) This begins with informal conversations to establish the social groupings of informants and select key representative informants of these groupings, then to using non-directive questions to allow informants to guide more detailed discussions as they see fit and highlight the issues they deem important, before checking emerging hypotheses with more directive questions as the relationship develops (Burgess, 1982; Smith, 2006).

The interviewing techniques adopted in this fieldwork also departed from this method in many ways. The methodology was directed and controlled by the researcher but it was essential to remain open to unexpected findings. Light (2007) shows that in the PRC ethnic categories and the rules of public identification are so artificially rigid that individual commitments cannot be reduced to ethnicity. One must then look outside these categories in order to understand Uyghurs’ multiple modes of political reflexivity, which often only emerge in jokes and asides. Roberts’ (2007) ethnography of a transnational Uyghur community based in Almaty argues that Uyghurs are state-less. Uyghurs do not feel a sense of allegiance to China or any of the Central Asian states in which they reside they are less susceptible to state-inspired hegemony (Roberts, 2007, p.204). In many ways, Uyghur-ness is born from daily interaction and practices, many of which are not tightly tied to cultural nationalisms or collective political action. In light of these considerations, additional open questions and free conversation were employed.

Allowing interviewees to talk about the topics they wished to before, after, and during the questions I asked enabled the research to be shaped by the topics which interviewees chose to raise such as the importance of food, language, and history. This flexibility allowed interviewees to self-select what social topics they deemed important and they could frame issues within the terms of reference as they saw fit. These adaptable techniques then allowed a flexibility in redesigning the project and a re-conceptualisation of “the problem” as it emerged (Thogersen and Heimer, 2006, p.3-4). In that sense we must integrate and even invite “serendipity” into the framework of our research (Pieke, 2000). One such example was that the original research design did not account for interviews with people who identify themselves as Han. However, through “serendipity” or the fact that they are less subject to “political repression” than Uyghurs, many Han were keen to tell me their stories (Anthony, 2011, p.51). Interviews with Han

63 became incorporated into the research and will be used, particularly in the final two chapters, to illustrate the problems the party-state faces in producing a singular national community based on ethnic unity.

The ‘progressively structured’ approach also allowed the opportunity for informants to exit the research situation. It was ethically important that those who seem unwilling to discuss specific topics or converse with the researcher must be left alone. Furthermore, due to the politics of the region, the first priority in any research project was the anonymity of interviewees. No real names or information were used in the recording or presentation of data. Coding was used for all names and key concepts at all times to record data. Note-taking took place shortly afterwards in private. All notes were stored in files with passwords on disks in locked, hidden storage. Some data was gathered without either the knowledge of the authorities or of the informants. This is justifiable in the production of knowledge within the context of authoritarian states because otherwise certain crucial methods of data collection are not possible at all (Smith, 2006, p.139). That would leave those who wish their voices to be heard as voiceless. Furthermore, employing this methodology allows respondents to stay out of political danger as they are not recorded as being involved in the research.

Data was collected through informal, detailed interviews in the Chinese language with approximately fifty Han and fifty Uyghur informants. Informants were all aged between 25 and 45. Approximately 25% of both groups were born in Ürümchi while the other 75% stated they moved to Ürümchi for opportunities in employment or education. This reflects Ürümchi’s position in Xinjiang as a hub for education and employment attracting both Uyghurs from other parts of Xinjiang and Han from other parts of China. The population of Ürümchi is approximately 80% Han in a nominally Uyghur majority region creating shifting majority- minority dynamics. It is a large, modern city where social interaction does not revolve solely around households or small neighbourhoods as Dautcher (2009) found in . The Chinese- built apartments mean that the “tradition” of Uyghur “open-doors” is not possible. Interaction then is either in public or in more private, enclosed apartments.

Interviews were conducted in Ürümchi between September 2009 and July 2010. Additionally, I was fortunate enough to conduct detailed, semi-structured interviews22 with three key Uyghur informants after building long term relationships based on trust. These interviews are considered in light of the social positions of the interviewee and compared with the broader data collected to address generalisability. Organising them will follow Joanne Smith’s (2000) adoption of Weberian ‘ideal types’ to suggest a degree of representativeness of particular

22 These were each about 6 hours long and conducted over several sessions with the informants I have the given the names Aynür, Mahigül, and Mukhtar.

64 categories of individuals. However, through their level of detail, the interviewees’ individual idiosyncrasies are taken into account. Despite the appearance of social rigidity, all symbols are mediated through the idiosyncratic experience of the individual and do not so much make meaning as give us the capacity to do so (Cohen, 1985, p.14). Furthermore, these idiosyncrasies are actively explored to gather a level of detailed data which cannot be obtained through quantitative methods. If we used a survey to ask Uyghurs if they are Muslim, the results would be close to 100%. However, this would tell us little about what being a Muslim means and how these meanings are negotiated in the politics of the everyday. The approach here employs “thick description” wherever possible to explore how individuals, culture, and social environments are embedded in one another in specific places and times (Geertz, 2000). However, due to the political obstacles, age range, , and gender are often all that can be revealed leaving the description thinner than it would be in more ideal circumstances.

In terms of how to select a sample of the Uyghur population, language was a key issue. Basic Uyghur language skills were invaluable to gain trust and engage with people. However, due to my own limited language skills, all interviews had to be conducted using Mandarin Chinese, with the exception of two using English. All practical matters thus pointed to a project based in the regional capital Ürümchi where almost all Uyghurs were able to converse at some level in Mandarin Chinese after gaining trust in Uyghur. There are limitations to generalisability on Uyghur self-identifications in a project based in Ürümchi using Mandarin Chinese. However, these limitations were adapted within the framework of this particular research project as strengths to focus on the regional capital, Ürümchi, where the Chinese nation-building project is at its most advanced. Using Mandarin Chinese and being located in Ürümchi means the research is focused on Uyghurs who have engaged thoroughly with the party-state’s nation- building discourses and can be seen as test cases for its effectiveness. This will explore how Uyghurs demarcate the boundaries of their self-defined identification, how they define what it means to be Uyghur, and how they understand the official discourses on Uyghur-ness and Chinese-ness.

Conclusions

Ethnic boundaries like nations are not givens but are in constant processes of production and (re)production. These processes are frequently contested and often through everyday social practices. Understanding the politics of identity requires an analytical framework which focuses on how the official and the unofficial interact and shape one another through this contestation of identity boundaries. This chapter reviewed the theoretical literature in the areas of security and nationalism before exploring their application to empirical cases of China and Xinjiang. It

65 outlined the theoretical and methodological approach of identity-security. The first section explored how security and identity are a means to articulate each other and demarcate the boundaries of Self/Other. The second section used literature from studies of nationalism and nation-building in China and Xinjiang to argue that nationalism is best understood as a process of identity-security performance from above and below. The third section reviewed literature from Chinese Studies and Xinjiang studies to understand how nation-building is popularly received and challenged by discourses of ethnic boundaries from below. The primary conceptual question in the thesis is how security and identity are inter-related in discourses of nation- building. However, this neither takes nations nor the articulators of nations as givens. Instead it seeks to problematise the boundaries of nation and ethnicity in Xinjiang through the concept of performativity. Identity and security are thus understood as discourses performed from above and (re)performed from below in official politics and the politics of the everyday.

This thesis will explore the limitations that ethnic boundaries in Xinjiang place on the production of a shared sense of multi-ethnic nationhood. It will examine how identities are securitised as a means to produce the boundaries nationhood but which inevitably produce security for some and insecurity for others. The approach explores how the conceptual and the visceral, state and people, official discourse and daily practices intersect, shape, and resist each other. These intersections are neither inherently harmonious nor conflicting and through contestation they produce complex, contextual, and relational Self/Other identifications. Do the official discourses on ethnicity and nationhood allow people to be incorporated into the Chinese nation while retaining their sense of individuality and existing group-ness at the same time? Is the securitisation of identity producing a secure Chinese community or are ethnic boundaries being (re)produced in ways that contribute to insecurity? To summarise in a single question, the thesis will ask what is China and how does it come to be thought of in this way?

66

Chapter 2: Writing the Nation: Performing a Timeless Self/Other

Introduction

This chapter will analyse the representations of the role of shaoshu minzu and Uyghur-ness as categories of persons within the official historical accounts of Xinjiang. The primary analytical concern is how these representations in official texts produce the boundaries and meaning of Minzu-ness and Chinese-ness. Nations and their symbols provide the capacity to make meaning and for collective action rather than guaranteeing the spontaneous emergence of groups from articulations of group-ness (Cohen, 1985, p.14; Brubaker, 2002, p.167-168). This chapter will judge texts on their capacity for identity creation amongst human beings by analysing the meaning of the identity categories offered by the party-state. What identity categories do they perform which people in Xinjiang must mediate and make meaning with? What type of nation does the party-state seek to produce through these historical accounts? The aim is to show how the boundaries of self-identification are articulated and securitised by the Chinese party-state in Xinjiang.

The study of history in Chinese Studies has seen the celebration of what Paul Cohen (2010) called a “China-centred perspective”. Cohen’s “China-centred perspective” takes Chinese sources and narratives as the starting point to understand Chinese history. Framing such sources as “Chinese” and taking ‘China’ as its unit of analysis may have been a useful remedy to the “impact-response” framework, which understood Chinese modernisation not on its own terms but as a passive response to the impact of the “advanced” and “active” “West” (Cohen, 2010, p.12). Academic Orientalism produced a generation of Sinologists studying China as an inferior form of the “West”. However, these charges can now be re-orientated and re-directed through studying Xinjiang where the region is represented as an inferior form of China. Cohen’s (2010) account reinforces the official party-state narratives analysed in this chapter. These official accounts can be understood as academic internal orientalism. By taking ‘China’ as their unit of analysis, this stretches the category back in time and distils competing regional, local, and often ethnic histories into a singular national history. Alternative historical interpretations and identities are excluded as being false as all people are forced to be included in Chinese history. This reversal of the orientalist charge underpins much of the analysis in this chapter. It is exemplified in Gardner Bovingdon’s criticism that all official historical accounts of Xinjiang written after 1949 are “China-centred” such that they undermine alternative accounts and centres, specifically those produced by Uyghurs (Bovingdon, 2004b, p.355). The exclusion of these accounts and centres is a crucial component of re-orientating Uyghurs’ self-identification from Xinjiang to China and Uyghur-ness to Chinese-ness.

67

Representations of identity are based on oppositional framings between subject and object and relationally defining social categories in terms of what they are not. This “constitutive outside”, in Winichakul’s terms, defines the national Self not by reference to any verifiable essence but by what it is not. This chapter then will explore how dichotomies of self/other, inside/outside, and security/danger mutually reinforce each other such that Otherness is what Judith Butler calls the “negative elaboration” of the Self (Butler, 1999, p.132). Through “negative elaboration”, the Other becomes an absence of the Self, performing the timeless boundaries of nationhood. The oppositional framing of self/other, inside/outside, and security/danger are mutually reinforcing dichotomies which reproduce each other. Where the Other is the absence of the Self, danger becomes the absence of security and vice versa. The reality presented in the texts for a domestic Xinjiang audience through political education is best understood as a form of nation- building. These texts are practices of identity management. The historical accounts used here project the past onto the present and onto the future. The dichotomies of self/other, inside/outside, and security/danger are represented as both timeless national realities and as pressing contemporary security concerns. Through these dichotomies these texts articulate what it means to be and what it must mean to be Chinese and an ethnic minority (shaoshu minzu). Meanings within each dichotomy are relational and mutually constitutive. These dichotomies then become mutually reinforcing as they are repeated and projected onto one another in the party-state’s nation-building discourse in Xinjiang.

How categories of identification are written into history is a central strategy of nation-building. “Nation-builders” see the creation of a “long and proud national historical consciousness” as a way to “construct nations and national identity” (Berger, 2007, p.1). National history secures for the contested and contingent nation the unity of a national subject through time (Duara, 1995, p.4). This “regime of authenticity” projects the territorial entity of the nation-state backwards in time, thus constituting an “unchanging core” moving through time and changing circumstances in a linear progression (Duara, 1998, p.288-291). Official Chinese histories of Xinjiang make the similar case that Xinjiang has always been an indivisible component of the Chinese nation and that ethnic Han have occupied the territory since antiquity (Bovingdon, 2004, p.355). The theoretical position of this chapter is that official texts on ‘national’ history reproduce a framed through bounded, timeless social categories, which humans mediate in order to comprehend reality. As Prasenjit Duara (1995) argued, scholars must “rescue history from the nation” and explore the contested and contingent meanings of history and community concealed by nations.

The chapter will be divided into three interlinked sections. The first explores the official historiographical narrative on how Xinjiang has been under the political jurisdiction of an unbroken, ethnic Han “nucleus” since ancient times. This section will examine how the party-

68 state writes the history of the Chinese nation in Xinjiang and the in the Chinese nation as inalienable but oppositional aspects of one another. This discourse appears across Xinjiang in publically displayed political slogans such as “Since ancient times, Xinjiang has been an indivisible component of the motherland” (Xinjiang ziguyilai shi zuguo buke fenge de yi bufen) and “Since ancient times, Xinjiang has been a multi-ethnic region” (Xinjiang ziguyilai shi duominzu de jujuqu). The second section explores how the party-state employs the historical division between pre and post-“liberation” Xinjiang, which contrasts “backward” feudal society against the modernity of the Chinese nation since 1949. This is an articulation of nationhood which depends on the division between Han and shaoshu minzu as relational and mutually reinforcing categories projected onto the feudal/modern dichotomy. This section will analyse how this dichotomy of Han-ness and Minzu-ness are officially conceptualised in relation to one another through the mutually reinforcing discourses of backwardness/modernity, frontier/neidi, and active/passive cultures.

The final section examines how the party-state projects history as a process of teleological unfolding from ancient history into the future. This teleology frames ethnicity in opposition to nationhood and as something which will be overcome through “ethnic extinction” (minzu xiaowang) and cultural “fusion” (ronghe). The argument is that there are at least four central dichotomies in these texts (Han/shaoshu minzu, modern/backward, active/passive cultures, and security/danger). These social categories do not reflect any fixed essence or reality. As representations they create the capacity for meaning-making and offer particular groups of humans the means to mediate reality. Within the internal logic of each dichotomy, meaning is relational and each side is mutually constitutive of the other. When these dichotomies are projected onto one another, they become mutually reinforcing. The Chinese nation is represented in Xinjiang as a unified, multi-ethnic, and developing community in contrast to its mutually constitutive textual opposition of the separatist, ethno-centric, and backward identification with specific forms of Uyghur-ness23. These dichotomies are ways of reproducing China as a culturally-defined nation and as a specific form of order; an ethnic hierarchy with the Han at the centre. This securitises a particular way of being Uyghur and being Chinese. It frames nationhood as a modern community and ethnicity as a backward threat, which in the name of security must be overcome.

23 Most of the material here is drawn from the contrast between the official representations of the categories of Han and Shaoshu Minzu in Xinjiang. This is because most of the official study materials on history and culture are framed this way. While Kazakh and Mongolian informants expressed disgust at these essentialisations they, like Uyghur informants, also expressed they felt these study courses were really directed at Uyghurs. The focus on Uyghur-ness in these texts will become more apparent in the chapters on performance (chapters 4 and 5) and when we move to discuss the criminalisation of Turkic identification (chapters 7 and 8).

69

2.1: Unifying Minzu in the Party-State’s History of Xinjiang, China, and the World

This section explores how the party-state seeks to unify all the peoples within China’s contemporary borders as timelessly minzu. During the communist period, the party-state and anthropologists stated that minzu theory is the “theoretical frontline” in “resolving the minzu problem” (Ya, 1965, p.224). It is thus with the political exigencies of minzu theory that this section begins. Contemporary debates on values in world politics often contrast the universalist human rights approach of the “West” against the cultural relativism of “Asian values” and “Confucian values 24 . The party-state frames commentary on human rights and appeals to “universal values” from outside China as an “excuse” by the “West” to “split China” (Ministry of Information, 2009, p.44-45). However, the party-state is concealing its own universalism in its policies towards ethnic minorities. The party-state’s approach here has always been that class and development are the “essence” of minzu identities. From the outset, this was a self- avowedly “universalist theory”, which applies to all of human history and its future (Shijian Bianji Bu, 1965, p.213). Minzu exist because class exploitation exists from earlier stages of development (Shijian Bianji Bu, 1965, p.213).

Today in Xinjiang, students of minzu tuanjie are similarly told that minzu separatism still exists because there remain “remnants of class exploitation” and “historical leftovers of old thinking” which “threaten national security” because they “want to destroy socialism” and “poison people” (XUAR Information Dept, 2009, p.84). This idea of universal stages of human development did not come solely from above. Anthropologists such as Ya Hanzhang wrote that “the essence of the minzu problem is a class problem does not only apply to a capitalist stage of development but also fits all historical periods prior to capitalism” (Ya, 1965, p.255). The opening of Ethnic Minority Policy tells readers that “the world we live in is a world of minzu” and that China is a “multi-minzu nation commonly created by all (Chinese) minzu” (State Council, 2009a, p.1)25. This goes some way to illustrating the importance of the category of minzu in how the party-state socially orders China and the world. Minzu is performed as a universal, objective truth and the boundaries between minzu identities are inevitable realities faced by all human societies. Particularism and cultural relativism only emerge through confrontation with alternative universalisms.

Minzu is framed as an identity that emerges at specific historical junctures as an obstacle to national unity but which modern nationhood will overcome. The “minzu problem” (minzu wenti)

24 For example, see contrasting reports on human rights by the State Council (Xinhua, 2012c) and the US Department of State (2012). 25 As with many of the themes presented in this chapter, this position is reiterated in most of the texts analysed, including Development and Progress, Xinjiang Cultural Knowledge, Common Knowledge, and The 50 Whys.

70 or how to combat ethnic separatism, is framed as something which can be tackled with the scientific discovery of facts. This is because Minzu separatism is not a Chinese-specific problem but a universal human problem, which emerge at “specific stages of socio-economic development” (Ethnic unity Education Board, 2009, p.26-27). This hierarchically juxtaposes tradition against modernity and ethnicity against nationhood. Ethnicity and ethnic minorities are thus framed as traditional and the Chinese nation as a modern means to overcome their tradition. Here we see mutually reinforcing role of the two crucial dichotomies of national/ethnic identity and modernity/tradition. Together these dichotomies form the official discourse of a nationally bounded and timeless social order.

The opening of Development and Progress, like most official statements on Xinjiang history26, establishes ancient Chinese historical jurisdiction and political authority over Xinjiang:

“Since the First Century BCE, the region of Xinjiang has been an important formative component of the Chinese nation. Furthermore it has played an important role in the building and development of a unified, multi-ethnic nation.” (State Council, 2009b, p.3)

Regions populated by ethnic minorities (minzu diqu) are logically essential to building a “multi- ethnic nation”. How Xinjiang is represented then plays a central role in what Benedict Anderson called the “imagining” of China as a “multi-ethnic nation”. The creation of a “long and proud national historical consciousness” which links Xinjiang to the rest of China is a central strategy of the party-state’s nation-building project. Every group of persons who have ever lived on the soil of what is today known as Xinjiang, “from ancient times until today”, are retrospectively and explicitly labelled as “self-identifying” (rentong) members of the Chinese nation (Zhonghua Minzu) (Ministry of Information, 2009, p.49-51). This is a political projection of modern onto ancient spatial domains:

From ancient times until today, many ethnic groups have lived on the territory of Xinjiang. Every ethnic group who has ever laboured, existed, and multiplied in Xinjiang has been a member of the Chinese nation (Zhonghua Minzu), including those minzu who once existed but are now extinct” (Ministry of Information 2009b, p.49).

Zhonghua Minzu is stretched back to a time before the very concept existed. This conflates imperial and modern domains in ways which naturalise the boundaries of the contemporary Chinese state. This historical narrative of timeless unity is cited as explanation of the contemporary Chinese territorial jurisdiction over Xinjiang and its peoples. The idea that the Chinese nation is timeless asserts territorial control (state-making) but also seeks to produce identification with the nation and the meanings offered by the party-state (nation-building). This timelessness draws social equivalence between the self-understandings of early stone-age

26 For example, see Xinjiang Cultural Knowledge, Common Knowledge, and The 50 Whys.

71 hunter-gatherers, who would still have been forming written language, all the way to present day ethnic minority students in Xinjiang who study modern, standardised Mandarin Chinese. They are all part of the Chinese nation (Zhonghua Minzu), a united community of self- identifying minzu. Nation-building then seeks to convert the living and the dead and absorb them into a single national community.

In many ways, historians and historiography in the PRC have been more effective than armies at incorporating Xinjiang into the Chinese state and severing its linguistic, cultural, and religious links to Central Asia (Millward and Perdue, 2004, p.36). This writing and performance of history is central to contemporary identity politics in Xinjiang. The party-state calls Identification with Turkic-ness “distortions of truth” and “distortions of history” because since 60 BCE “every minzu identified with the command of the central rulers…and even now-extinct minzu were members of Zhonghua Minzu (Ministry of Information, 2009, p.47 & 49-50). Not only does this framing of history imagine a community with the dead to create the unbroken national history. It stretches nation-building back in time to absorb and convert those peoples referred to as barbarians (yi) in Chinese texts but who called themselves Turks. This discourse articulates Uyghurs and Uyghur history as latently Chinese in a process of teleological Sinicisation. The 50 Whys tells readers that Uyghurs are “not a Turkic minzu” because “after the Turk collapsed in the 8th century, they did not form a modern minzu” (Ministry of Information, 2009, p.57). The Turk is thus performed as an extinct Chinese group. This denies modernity to the identification with the category of Turk because they did not form a modern minzu, which relationally produces Han-ness as an unbroken lineage at the core of Chinese-ness. The idea that there is such a thing as Turkic identification is a “myth fabricated by Pan-Turkism” (Pan, 2008, p.12). An historian would note the conflation in this narrative between the Gok-Turk kingdom, which did collapse in the 8th century and the broader category of Turk, which has been employed in nationalist movements throughout the 20th century in Turkey and across Central Asia. However, the function of these texts is less to represent reality than it is to produce a reality, which offers politically acceptable models of self-identification within a larger nation-building project.

From the outset of one party-rule, minzu was institutionalised as an ethnotaxonomy mapping the ethnic makeup of China through its classification project (minzu shibie) (Mullaney, 2011, p.135). The first task of minzu shibie was to differentiate which groups were Han and which groups were shaoshu minzu (Fei, 1980, p.166). This exemplifies the social practice of the Han/Shaoshu minzu dichotomy from the outset of party-state rule, which built on rather than revolutionised the distinction between Hua and Yi. Marxist-Leninism and Stalin’s four principles of nationhood 27 are cited as the basis of the party-state’s ethnic minority policies and the

27 These four principles are common language, common territory, common economic life, and common psychology. See: Stalin, Josef (1913/1945).

72 solution to the minzu problem (Pan, 2008; Hu, 2009). However, the party-state departed from Stalin and Engels by rejecting the idea that nation and ethnicity were intimately connected to modernity and a capitalist stage of development (Ya, 1962, p.117-119).

The party-state worked to re-write Marxism so as to fit Marx to Chinese history as much as Chinese history to Marx. In the spring of 1962, the party-state held a conference in Beijing to “unify” translations of Marx, Engels, Lenin, and Stalin into Chinese. The conference decided to abandon all use of (buluo) and (shizu) thus translating all social groupings in Marxist works from the beginning of time and forever as minzu (Ya, 1962, p.125-126). This was a “regime of authenticity” to secure a fixed national subject moving in a linear progression (Duara, 1995, p.4; 1998, p.288-291). Political control over the Marxist conceptual framework, so central to political vocabulary in China, has the effect of erasing the self-identified differences between groups from the historical record. All and become ephemeral transitions to the inevitability of unification under the minzu category. Culture is then understood as a teleological unfolding of minzu. The concept is stretched back to frame groups as minzu from times before the word existed. All groups who have existed or indeed become extinct on the territory today we call ‘China’ retrospectively become Chinese such that their contemporary descendents are equally unable to escape this identity-teleology. ‘China’ is also stretched forwards as its applicability to every period of history means humans will be minzu, at least until minzu withers away under Communism. This teleology simultaneously articulates belonging and difference because the “fixed national subject” has an internal structure where national subjects are differentiated according to their minzu. In China, history ought to be rescued from the nation and from the minzu category.

“Semantic hybridity” will be explored in later chapters but here it is prudent to note that ideas about Confucianism, socialism, and ethnicity in China do not always fit comfortably together with logical ease. However, they often do reinforce one another. The party-state and Chinese anthropology have taken the teleology of Marx-influenced universalisms and transplanted it onto what James Leibold (2011) called “Confucian ecumenism”. The timeless Chinese civilisation includes “barbarians” by converting them to civilisation but now also to modern nationhood. Minzu can be retrospectively applied to every period of history but there are “backward minzu” and “civilised minzu” (Ya, 1962, p.116). Ethnic unity or minzu tuanjie28, the national bonds which are said to tie different minzu together, similarly flows from the dawn of human beings in Xinjiang all the way to life in the contemporary nation-state, which we call China. These claims are explicitly about the building of the nation as a timeless, self-identified community rather than simply the mapping of the military control of state-making. Xinjiang has

28 This concept and its relation to “The Three Evils” will be analysed more fully in chapter 5 because of their importance in the criminalisation of self-understandings and discourses on identity which challenge the official narrative.

73 not simply to be understood as a territorial component of China but its peoples must identify with and desire to be part of China. Every Shaoshu Minzu in Xinjiang is said to have welcomed their “liberation” from their “backward condition” with the arrival of the PLA in 1949 (Ministry of Information, Theoretical Department, 2009b, p.150-151).

To pass exams on these texts, individuals were required to link their own self-identifications to these particular perspectives on history through repetition of its social framings. Several students interviewed explained how they had to answer exam questions on these texts such as “why do we say that Xinjiang has always been part of the Chinese nation (Zhonghua Minzu)?” and “why do we say that by 60 BCE, Xinjiang was already an inalienable part of the motherland (zuguo)?”. This political education and its status as a requirement to success in education is a strategy to tell people they must understand themselves and their ancestors as members of the Chinese imagined community.

The concept of shaoshu minzu appeared in Chinese for the first time in the 1924 CCP and GMD as difference between humans was reconfigured from distance from the Chinese centre to a series of bounded national units (Leibold, 2007, p.9-11). The ostensible sameness inherent to national community conceals the ambiguous position of Xinjiang within Zhonghua Minzu. Xinjiang is only referred to in the political education texts and the white papers in terms of its incorporation into Zhonghua Minzu. Xinjiang’s entrance in this narrative begins with the Chinese entrance into Xinjiang and the teleology of the unification of Zhonghua Minzu. In all of these official texts, the account opens with the ’s (206-221 BCE) establishment of “jurisdiction” over Xinjiang understood as a “frontier” (bianjiang) and as an “ethnic minority region” (State Council, 2009a, p.4)29. These exclusionary ascriptions of Xinjiang are repeated for each period of history thus reinforcing the timelessness of Xinjiang’s frontier- ness. The terms are not specifically defined but the reader is assumed to know what they mean as frontier and centre remain powerful symbols demarcating the internal cultural boundaries of China. The centre of official narratives on Xinjiang history is not Xinjiang but the central plains of China such that Xinjiang is peripheralised as a place to be conquered and pacified (Bovingdon, 2004b, p.355). Pre-modern boundaries determined by distance from the centre continue to be performed by the party-state but largely contained within the boundaries of modern borders30.

The official historical narrative then characteristically leaps more than eight centuries from the Han dynasty to the (618-907 CE) when again we are told Chinese “jurisdiction”

29 Also see: State Council (2003) p.52; Ministry of Information, Education Department (2009) p.1-2; Ministry of Information, Theoretical Department (2009b) p.149. 30 Chapter 6 will show how these pre-modern conceptualisations of space emerged in interviews and discussions with ordinary people on a daily basis through the idea of Xinjiang as a frontier.

74 included Xinjiang (State Council, 2009a, p.5) 31 . What is written here is perhaps of less importance than what is not written. Those dynasties which represent discontinuity in Chinese control are not written into this narrative so as to represent Xinjiang as timelessly Chinese. The narrative conspicuously omits an eight century period, known for the absence of Chinese presence and indeed the rise of Uyghur city-states such as and Jiaohe, which continued to exist during the Tang era (Millward, 2007). The Party-State’s narrative then leaps again, this time to the (1206-1368 CE) and then to the (1644-1911 CE). These sections tell readers the Qing “guaranteed” the present day borders of China and the rule of centralised government, which had been established during the Han Dynasty (State Council, 2009a, p.5)32. What little mention is made of intervening periods such as the Song, Liao, and Jin dynasties, readers are told that war on the Central Plains meant that the Chinese state was too busy to “look after” Xinjiang (Ministry of Information, Theoretical Department, 2009b, p.149-150). Xinjiang’s position in China is then articulated as one of timeless dependence. It is performed as an inalienable part of China but its difference is reinforced through its inclusion as a dependent and distant frontier. Readers are then immediately reminded that close but unspecified political links were maintained and that unity and national identity crosses historical periods all the way to today (Ministry of Information, Theoretical Department, 2009b, p.150-151). This historical narrative may be as partial and politicised as any other alternatives in the region. However, it has a political apparatus to establish it as singular truth by vociferously promoting it in all education and political institutions.

Historians such as Peter Perdue (2005), Laura Newby (2007), and James Millward (2007) tell us that the Qing Dynasty integrated the whole of the territory of what is today known as Xinjiang for the first time into the Chinese state. Xinjiang’s history is better understood in terms of its “in between-ness”; between the Chinese, Islamic, Indian, and Mediterranean worlds, none of which predominated the whole of Xinjiang and none of which predominated for extended periods (Millward, 2009, p.55). However, the history of Xinjiang is being officially written and disseminated by selecting specific periods of Xinjiang history when Chinese troops are present and neglecting others when they are absent or less influential. The history of Xinjiang is written as a history of military conquest where the presence of troops is equated with the unification of nationhood.

Chinese unity and national community are represented as timeless and inevitable but which, over time, grow in strength. The “overall direction” of history has always been towards “national unification” and “all minzu took the establishment of central Chinese authority as

31 Also see: Ministry of Information, Education Department (2009) p.2; Ministry of Information, Theoretical Department (2009b) p.149. 32 Also see: Ministry of Information, Education Department (2009) p.2-3; Ministry of Information, Theoretical Department (2009b) p.149-150.

75 national orthodoxy…today, Zhonghua Minzu has already become the common name and ascription for the universal self-identification of every minzu” (State Council, 2009a, p.5 & 7). The period since 1949 under the CCP is but the highest stage reached thus far of ethnic unity, national prosperity, and common identification with Zhonghua Minzu (State Council, 2009a, p.7)33. In this way the party-state and its historical narrative are legitimised as representatives of the most advanced stage of history but also as bearers of the continuity of an unbroken national lineage. This enables China to be both modern and ancient at the same time.

All the official texts selected here refer to China as a “unified and pluralistic nation”, which is borrowed from the late anthropologist Fei Xiaotong. Fei (1988) claimed the formation of the Chinese nation since ancient times revolved around the developing, culturally superior force of the Han “nucleus” (Fei, 1988, p.214-215). Fei tells us that the formation of the “unity” and “plurality” of the Chinese nation was formed by the Han who engaged in agriculture on the central plains of the lower reaches of the Yellow and Yangtze Rivers (Fei, 1988, p.168). The Han nucleus has since ancient times driven economic development and the unification of the Chinese nation. “Plurality” refers to the plurality of groups, now referred to as minzu, who have been absorbed into and modernised by China through conversion and attraction rather than conquest (Fei, 1988, p.215). The dichotomies of backwardness/modernity and frontier/neidi are then mutually reinforcing aspects of the same discourse of ‘China’. Minzu retain some ethnic characteristics but have become Chinese. Conversion is then said to be unlike the military of the “West”. It is said to be non-violent and an inevitable outcome of coming into contact with the irresistibility of the superiority of Chinese culture. Difference is then inscribed at home and abroad at the same time. Pluralism in China then is as much about the ongoing existence of plurality as it is about the absorption of plurality into the Han nucleus.

Following Fei Xiaotong, the official narrative represents Han-ness as an ancient and unbroken lineage. Xinjiang Cultural Knowledge describes Han culture as an ancient lineage, which runs unbroken to the present and is fused from both an ancient culture (Han wenhua) and a modern one (Hanzu). Han culture (Han wenhua) as the postulated basis of modern Han identity is said to have risen during the Han Dynasty and flourished during the Tang. This account of Han-ness contrasts with accounts of Minzu as a modern political construction of the Chinese state. Conversely, the lineage of Uyghurs is framed as broken through the “extinction” of the Turks (Tujue) and Huihu/Uighurs (Huihu) (Ethnic Affairs Commission, 2009, p.8-9). Huihu is a transliteration of Uighur, the ethnonym for the group who migrated from the Orkhon valley in present day Mongolia to the Turpan region in the 9th century. However, in Chinese academic literature it is always translated as Uygur and never Uyghur to avoid linking the two. This

33 Also see: State Council (2003) p.54; Ministry of Information, Education Department (2009) p.5; Ministry of Information, Theoretical Department (2009b) p.150-151.

76 conceptual separation of ancient from modern ethnic categories, despite the origin of the term Uyghur, conceals the ongoing identification with the category of Turk in the region34. Han-ness is being relationally defined as the timeless, advanced essence of the nation against the feuding tribal politics of minorities, which have been united by and through the superiority of the Han. Fei, like all the official texts, conceptualised the Han as the core of five thousand years of “nation-building” where the history of the Chinese soil is one of inevitable development, unification, and “assimilation” into Zhonghua Minzu (Fei, 1988, p.219). Official narratives of Chinese history are told through the terms of the present to prove the inevitability of history and thus the social world today. This is as opposed to telling history from the perspective of the past to illustrate the different directions history could have taken. This closes the possibilities to imagine the past and how it shapes the future, for there can be only one past (Shaoshu Minzu) and only one future (Zhonghua Minzu).

The sameness in representation for all different ethnic groups under the bracket of Shaoshu Minzu conceals a hugely diverse range of linguistic families, religions, and trans-regional histories. Population figures, historical experiences, and economic features are continually referred to in this text as either Han or shaoshu minzu (State Council, 2009a, p.2-3). A huge range of histories are absorbed in the category of nameless Shaoshu Minzu defined as opposition to and absence of Han-ness. Lacking any objective essence, these categories only have meaning in relation to each other and the emptiness of the category Shaoshu Minzu is as much about providing a sense of self to the category of Han as it is about nation-building to include other ethnic groups. That urban political and social centres such as Gaochang and Jiaohe flourished in the medieval period of Xinjiang history has no place in this narrative. These representations cannot be understood by comparing them to ‘reality’. They only make sense within the series of mutually reinforcing dichotomies which form a powerful and coherent discourse on what it means to be Chinese.

The Chinese nation as a cultural hierarchy is being produced with the Han at the centre and Uyghurs on the periphery or the frontier. This divides China into Han/minority, centre/frontier, and modern/backward, exemplified in Fei Xiaotong’s (1988) vision of China as a pluralistic, unified nation developed by the Han core’s cultural and economic superiority. The dichotomising approach to Han/Shaoshu Minzu and nation/ethnicity is reinforced through party-state’s discourse on “backwardness” (luohou) and “liberation” (jiefang). The official texts on the party’s ethnic minority policies justify themselves through and embed themselves in this narrative of backwardness and liberation which come to define Minzu-ness and Han-ness. This binary

34 This thesis does not seek to verify or falsify links between the ancient and modern with regard to Han or Uyghurs. The analytical concern here is how the asserted presence and absence of such links are employed by the party-state in these nation-building texts. Chapter three will offer fuller analysis of the use of the term “Turkestan” in Xinjiang.

77 approach reproduces ‘Chinese’ identity as a historically ancient, nationally bounded, hierarchical community where, as we shall see in the next two chapters, alternative self-identifications are represented as security threats to its unity and prosperity. Ethnic Minority Policy tells us that prior to liberation, shaoshu minzu engaged in herding, hunting and fishing in grasslands, deserts, forests, and various other nameless natural habitats (State Council, 2009a, p.6). Defined as backward, these regions are then contrasted against the relatively modern agricultural economy of the central plains and Han-populated regions (State Council, 2009a, p.6; 2009b, p.4; Ministry of Information, Theoretical Department, 2009b, p.151-152). What Judith Butler (1999) called “negative elaboration” is practiced by the party-state in its ongoing project to articulate the internal boundaries of China where Han-ness is represented as modern and active against the backwardness and passivity of Shaoshu Minzu.

How to frame representations of Minzu and nation in official historical accounts is fraught with essentialist pitfalls. These narratives are not told from the perspective of the pluralistic groups of peoples who have formed the Chinese nation but seemingly from the point of view of the Han “nucleus”. This nucleus is said to have incorporated the minority Other and wrote them into ‘Chinese’ history today. However, to call it “China-centred” assumes that there is some objective, uncontested, non-contingent China to which it refers. Similarly, if one succumbs to the theoretical convenience of framing it as Han-centred, would open us to the charges of essentialising Han-ness, an equally contested and contingent category of self-identification (Chow, 2001, p.47-48). The major contributions to understanding China from Dru Gladney (2004) and Louisa Schein (2000) was to show how shaoshu minzu are peripheralised but not peripheral. Representations of minorities are central to the relational production of Han-ness and this peripheralisation is crucial to providing a centre. As discussed in chapter 1, the literature of the party-state, including the constitution, employs a binary division between Han (Hanzu) and “ethnic minorities” (shaoshu minzu). In the same way that the constitution refers to Han and minorities, as if minority is a form of self-identification as opposed to Uyghur or Tibetan or Mongolian, Han/minority and frontier/neidi are essence-less dichotomies which mutually reinforce one another. Non-Han groups are defined “positively” according to cultural practices, but “negatively” according to a lack of Han Chinese attributes in a binarising discourse of hierarchical Chinese identity politics (Callahan, 2010, p.132). Minority is merely the absence of Han-ness and the frontier or ‘ethnic minority regions’ are simply the absence of neidi. These categories are ways of articulating each other.

Despite the best intentions of the authors, the relational framing of frontier/neidi and Han/Shaoshu Minzu within these texts can help us see how these objects have no uncontested essence or reality independent of each other. It is then better to critique this narrative as centred history. It is its essentialisation per se of individuals bounded into social categories

78 which is problematic. These essentialisations project the contemporary boundaries of politically ascribed categories of self-identification onto the past and close social possibilities in the present. This closure seeks to prevent different persons, as participants in discourses of identity, from imagining themselves in ways other than through the political lens of Zhonghua Minzu, defined as a Han nucleus around which shaoshu minzu revolve. These essentialisations seek to speak for Han and non-Han35 alike albeit in very different ways. These essentialisations are ways of performing China as a social hierarchy, which defines Han-ness as presence and minzu- ness as absence.

2.2: Producing Minzu-ness, Producing Han-ness

The constitutive elements of the Chinese nation are best understood through “the extremes of modern/traditional, barbarian/civilised, marginal/core, and inside/outside” where China is produced somewhere in between (Barabantseva, 2011, p.276). This section is focused on how these stark dichotomies reproduce difference between Han and Shaoshu Minzu and how this difference is incorporated into the discourse of a timeless Chinese nation. As one of the most successful and politically influential theorists of ethnicity and nation in China, Ma Rong (2007), a scholar of Hui ethnicity, argues that it is culture and not ethnicity which define social distinction in China. The distinction between “civilisation” and “barbarians” in ancient China and different minzu in contemporary China then orders the nation (Ma, 2007, p.7). The barbarian/civilisation distinction is not understood as a distinction between different civilisations but between “highly developed and less developed ‘civilizations’ with similar roots but at different stages of advancement” (Ma, 2007, p.5). This highly influential framework positions identification with China as positively correlated with economic development such that ancient “barbarians” naturally became Chinese by being developed and thus, learning Chinese culture (jiaohua). Li Dezhu of the State Ethnic Affairs Commission similarly stated in 2000 that minority nationalities need to learn from the advanced culture of the Han. This privileges Han Chinese identity as containing a “civilising imperative” and characterises the ideological foundation of “modernisation” (Goodman, 2002, p.137; Goodman, 2004a, p.326-329). This framework orders the Chinese nation as bounded into groups who are more and less Chinese and more and less modern as opposed to different.

Adopting an implicitly ‘Han’-centred voice, Ethnic Minority Policy refers to “surrounding ethnic minorities”, placing the central plains and the Han at the centre and other ethnic groups on the outside. The text continues, telling its readers that since ancient times these nameless “surrounding ethnic minorities” and the “central plains region” promoted a “mutually

35 Non-Han here is not used to refer to a form of self-identification but a form of official representation. It is applied to all ethnic groups other than the Han and defines them through the absence of Han-ness.

79 complementary economy” (jingji hubu) and “common development” (gongtong fazhan), which formed the basis of the unified, multi-ethnic Chinese nation (State Council, 2009a, p.6). This internal orientalism excludes Shaoshu Minzu from what is posited as an ancient, advanced cultural core but then includes it as part of an asymmetrical but “complementary” economic whole. Ethnic Minority Policy explains that Shaoshu Minzu provided the central plains with goods in return for their continued existence:

The surrounding ethnic minorities and the central plains region engaged in a ‘mutual tea-horse market’……as well as satisfying Central Plains agriculture, transport, and military needs for horses, it also supplied Shaoshu Minzu with everything they need for daily living (richang shenghuo suoxu), advancing a complementary economy and common development. (State Council, 2006, p.6)

The 50 Whys discusses this complementarity in similar terms of the production of a unified, multi-ethnic nation formed through holistic relations of dependence between advanced national core and backward ethnic periphery. The function of shaoshu minzu provision of horses to the central plains is said to be the development of their own living areas (Ministry of Information, 2009, p.66). This is complementary to the Han provisions of silk, tea, and daily necessities, the function of which is theorised as to “promote the development of ethnic minorities and ethnic minority regions” (Ministry of Information, 2009, p.66). A hierarchical representation of the Han-Shaoshu Minzu relationship emerges here. Ethnic minorities have provided the Han with horses but this is a supplementary requirement for further development of Han regions. The inclusion of Shaoshu Minzu into China on the other hand, is written such that since ancient times minorities have needed the Han not only for development but for their very existence; “everything they needed for daily living”. Through this relationship of dependency, the nation is written as an ethnic hierarchy where all ethnic groups are included but the Han are the nucleus.

Xinjiang Cultural Knowledge categorises the cultures of Xinjiang into material divisions of Oasis animal husbandry, Grassland Nomadic, and “Frontier Culture” 36 (Ethnic Affairs Commission, 2009, p.8)37. These multiple poles are said to have fused to form the distinct culture of today which exemplifies the region and has always been a part of Chinese culture (Ethnic Affairs Commission, 2009, p.23 & 134). This explicitly materialist account of culture defines identity through the concept of mode of production (Ethnic Affairs Commission, 2009, p.2, 4, & 8). Uyghurs are defined through animal husbandry, and Kyrgyz as nomadic herders, and Han as “builders”. The customs, religion, and “modern formation” of Uyghurs and all non-Han groups in Xinjiang are explained as superstructural to their material “base” and subsequently

36 This would be literally translated as “station troops to open up the wasteland culture” (tunken wenhua). 37 These categories are referred to in other Xinjiang-specific official texts but not in those disseminated nationally. These books are read by individuals of all ethnicities in Xinjiang so it reflects the local need to acknowledge the existence of groups within the shaoshu minzu category.

80 categorised as “oasis” or “nomadic” (Ethnic Affairs Commission, 2009, p.24-55; 56-79). This is an evolutionary cultural framework in which culture is defined through the mode of production and as this mode develops from savagery to civilisation so too does culture. This theory of cultural evolution is not confined to official texts as the pioneers of this essentialist and materialist approach, Lewis Morgan and Friedrich Engels, are top of leading regional universities’ anthropology reading lists 38 . Friedrich Engels (2010) explained power relations between the sexes, such as and , through social relations embedded in the mode of production. This text is often cited in Chinese anthropology to explain the matriarchal structure of the of Yunnan as a “living fossil” from a pre-, primitive communist mode of production (Walsh, 2005, p.456).

The party-state’s approach to ethnic minorities is that development will resolve “national contradictions” such as the “ethnic problem” (minzu wenti) as all people are unified into Zhonghua Minzu (NPC, 2001; Hu, 2007, p.13-14; Bekri, 2008). Hu Jintao’s concept of “Scientific Development” (kexue fazhan) is usually explained by reference to another slogan: “people- centred” (yirenweiben) (Ministry of Information, 2008, p.26). This refers to the development of people including education, Chinese language literacy, income levels, and overall “quality” (suzhi). It also positions itself as “scientific Marxism” to emphasise the “objective” approach of the party-state based on “truth” and “facts”. When applied to people in Xinjiang, this also denotes an evolutionary cultural framework, with identity understood as superstructural to economic development, as the basis of the party’s ethnic minority policies and its “ethnic minority work” (minzu gongzuo) (Hu, 2007, p.13-14; Ministry of Information, 2008, p.26). This “Scientific Development outlook” (kexue fazhan guan) is understood by the party as a matter of national security and stability because it will resolve “national contradictions” such as the “minzu problem” (minzu wenti) and ensure the unification of identity around the party within the present boundaries of China (NPC, 2001; Bekri, 2008). This securitises modernisation and identity such that the teleological flow of history towards the resolution of contradictions and shared identification of Zhonghua Minzu is framed as security against the threat of backwardness and the minzu problem.

This discourse of cultural evolution is performed in ways which build on rather than supplant identity politics based on lineage and Han ethno-centrism. “Han chauvinism” was identified by the CCP as a cause of “minzu alienation” (minzu gehe) but emerging through class exploitation (Shijian Bianjibu, 1965; Ya, 1965). However, narratives on Han culture, defined as “frontier

38 This information was drawn from interviews with students and teachers at Xinjiang University and Xinjiang Normal University. Lewis Morgan’s Ancient Society (1944) remains a key text and is celebrated in Chinese anthropology (eg see Pan Jiao (ed) (2008)). His theory, which was discussed in chapter 1, was that culture can be explained in terms of three stages of evolution from savagery, to barabarism, to civilisation. For example, see: Morgan, Lewis (2005) Ancient Society and Engels, Friedrich (2010) The Origin of the Family, Private Property, and the State.

81 culture” in Xinjiang today, do not draw the same link between mode of production and identity. It frames Han-ness as its active contribution to building the frontier and the nation with their active, advanced culture raising the cultural and technological levels of Xinjiang (Ethnic Affairs Commission, 2009, p.13, 18-19, & 80-82). As pioneers and frontier-builders, the Han are not simply placed at the top of the evolutionary cultural framework but they are placed outside it. Han-ness as a representation embodies the true communist spirit of man’s domination over nature by transcending both nature and ethnicity through nationalism. Han-ness, as the nucleus, embodies the spirit of the nation and of modernity. Han-ness is the active force which drives the Telos of Chinese history. The pioneering “frontier culture” as the core of the nation in both ancient and modern times, is represented as a lineage beginning with the military incursions into Xinjiang during the Han dynasty and running to the contemporary presence of the Han in Xinjiang today as force for “modernisation” and cultural development (Ethnic Affairs Commission, 2009, p.80-82):

Through the written record and cultural artefacts found in Xinjiang, as well as frontier relics of the Han and Tang, it is impossible not see that Han culture has from beginning to end co-existed with all ethnic groups in a state of fusion and complementarity. In every aspect it has promoted the development of the culture of the Western Regions. (Ethnic Affairs Commission, 2009, p.81)

The “frontier culture” of the Han is represented as that which actively drives the progression of history through development through its introduction of advanced technology, the earliest form of writing in the Tarim basin, and patriotic poetry (Ethnic Affairs Commission, 2009, p.81-83). In this narrative, the Han are building the Chinese nation for all minzu. This relationally represents groups other than the Han as passive to nature and subject to the progression of history. This narrative includes Xinjiang Shaoshu Minzu into China but it excludes them from modernity and from the national nucleus by defining them through a relationship of dependence with the Han. Elena Barabantseva’s (2011) Decentring China showed how the party-state’s development discourses frame ethnic minorities as localised elements of the nation-state thus excluding them from equal participation in debates on modernisation. The party’s representation of the minorities as “ethnic” and “backward” in contrast to the advanced, mobile Han Chinese populated East reproduces a hierarchical socio-spatial order where the “backward”, “ethnic” West must learn from the advanced, mobile Han Chinese populated East (Barabantseva, 2009, p.250). As a category defined through the natural environment and through the mutually constitutive presence of Han-ness and modernity, Shaoshu Minzu helps produce a hierarchical nation, without which the category could not exist.

Ethnic minorities are included in Zhonghua Minzu based on the advanced culture of the Han brought to their “frontier” regions. The party’s discourses on the development of ethnic minority regions tell a narrative of history which begins with “peaceful liberation” (heping jiefang) by the

82

Red Armies in 1949 (State Council, 2009b, p.4). The 50 Whys explains that the party and the Red Armies peacefully “liberated” Xinjiang from its “backward” history of being “without tall buildings” (Ministry of Information, 2009, p.40 & 79). Ethnic Minority Policy tells readers that the new China’s mission was and remains to build from an “extremely backward” base of traditional agriculture and “low productive power” (State Council, 2009a, p.24-25). The party- state’s usage of “frontier” echoes Frederick Jackson Turner’s (1986) classic definition of the American frontier from 1893 as the “meeting point between savagery and civilization”. This imagines a frontier as outside the self but a place on the fringe where wilderness and savagery can be transformed into a liveable and civilised place (Smith, 1996, p. xv). The progress and development of Xinjiang is framed through identity politics as “the result of the common struggle of the people of every minzu to raise high the banner of minzu tuanjie” (State Council, 2009b, p.3). Due to the “persistence” (jianchi) of the nation, the “struggle” for ethnic unity, and the “preferential policies” (youhui zhengce) of the party-state, minorities are said to be entering their most prosperous period in history (State Council, 2009a, p.25). The colonial undertones of this “persistence” are not restricted to official politics and was popularly captured in the hit television series “8,000 Flowers Go up the Mountain” (baqianxiang shangshan). The show was produced by Television (Hunan Dianshitai) and interspersed Television drama with interviews with the “Hunan Flowers”. It told the story of the transfer of 8,000 young girls from Hunan to Xinjiang after “liberation” in order to provide wives for the newly settled Chinese soldiers. In the opening episode the leading Communist General gives a rousing speech explaining that the reason the Chinese military have to once again “liberate” Xinjiang is because the corrupt Qing dynasty did not “persist” thus they failed in their historical duty to unify the nation.

Development and Progress similarly describes the “post-liberation” period, particularly the reform period as one of rapid development and “Xinjiang building” under the leadership of the party-state (State Council, 2009b, p.4). Xinjiang is then marked out as a region not simply in need of nation-building but its own particular project of “Xinjiang building” to maintain its position as a site of internal orientalism. “Pre-liberation” Xinjiang is understood as a “natural economy”; a “backward”, “stagnant” region “without technology” and almost entirely “illiterate” (in Mandarin Chinese) (State Council, 2009b, p.4-7, 19-21). After reading these representations of Xinjiang as a place incapable of offering security or prosperity, the central party-state enters the narrative as benevolent provider of more than 50% of post-“liberation” investment in the form of subsidies. These have led Xinjiang to its most prosperous period in history (State Council, 2009b, p.10). Furthermore, the party has sent “experts” from “Inner China” (neidi) to Xinjiang in order to facilitate the “study” of both the technology and the “” (guannian) of Inner China (State Council, 2009b, p.11). The party-state then elaborates its role on the

83

“frontier” not simply as an economic provider but the continuation of an historical role to facilitate the development of thought which corresponds with (how it understands) Inner China.

“Modernisation” is not simply about material or economic process and in Xinjiang we can see how it can be ethnocentric and deeply ideological. Modernisation in Xinjiang is intimately related to transforming how people identify themselves and can only be understood within discourses on Minzu and nationalism. It represents Xinjiang as an indivisible region of the Chinese nation. However, it is one which is a backward “frontier” and dependent on the centre for its very survival in the same way Shaoshu Minzu are represented as dependent on Han for their existence. “Pre-liberation” as a category of framing history then emerges to relationally define all history prior to 1949 as backward and passive in opposition to the modernity and active nature of “post-liberation”. This dependency narrative of passive, “backward” minority regions being “modernised” by the party is pervasive throughout these texts and is the contemporary continuation of 5,000 years of ‘Han’-centred nation-building. The narrative is told chronologically beginning with the ancient economic relations of the horse and tea trade, running to the policies of the contemporary party-state. Such a narrative aims to produce a “long and proud national historical consciousness”. This consciousness is based on ethnic hierarchy with the party-state represented as the inheritors of an unbroken national lineage of cultural and economic advancement. The meaning of Han and Shaoshu Minzu are being produced through the hierarchical and mutually constitutive relationship between the two categories. What is particularly problematic here is that these performances of Self/Other exclude shaoshu minzu from modern nationhood by marking them as culturally inferior to the Han nation-builders. However, this is also how shaoshu minzu are being included. They are included through the active building of the Han, whom they should identify as part of the same nation but also as superior group building the nation for everyone. Zhonghua Minzu, thus understood, is based on internal orientalism. Difference on a qualitative level is not acknowledged as such; minorities are merely quantitatively behind the Han but can be developed and converted.

2.3: Overcoming Minzu with Han-ness

This section is about how discourses of minzu are ultimately aimed towards their own destruction through the teleological unfolding of minzu extinction (minzu xiaowang). While minzu is represented as a timeless category, the expectation of the party-state has always been that minzu identities will fade away first into Chinese national identity and later into a classless society without the need for social distinction. This section will analyse how these ethnicisations mark off Xinjiang as a particularly ‘ethnic’ and dangerous place. These ethnicisations are part of a discourse on how modernisation overcomes ethnicity to produce nationhood as a form of

84 security. It will show how “Fusion” (ronghe) and “multi-culturalism” (duoyuan wenhua) are intimately linked to minzu extinction within this framework of transformative identity politics. These discourses draw on the wealth of historical representations we have analysed in order to project a vision of the future without Minzu.

The modernisation and the understanding of Chinese-ness the party-state offers are securitised in Xinjiang such that alternative self-identifications are represented as threats. Ma Dazheng, the influential head of the Frontier Research Institute in Xinjiang, stated in a document for internal circulation only that “Hans are the main source of stability for Xinjiang” (Bovingdon, 2010, p.58). In 2008 , the chairman of the regional government, announced that those who challenge the party’s modernisation policies are equivalent to terrorists, “backward”, and “crazed” (Bekri, 2008). The Western Development Project and the Ethnic Question by Yang Faren of the Xinjiang Regional Party Political Research Office, similarly ties security threats and non-Chinese self-identifications to backwardness and levels of cultural and economic development. Yang argues that separatism only appears in less developed, rural, ethnic minority regions. The modernisation of the Western Development Project (Xibu Da Kaifa), including “bilingual” education policies, is then represented as the promotion of political stability through the development of minorities’ material and spiritual civilisation Yang, 2004, p.1-3). This is to be achieved by attracting high ‘quality’ (suzhi) talents from Inner China to Xinjiang from whom shaoshu minzu can learn modernisation (Yang, 2004, p.1-3). Within this discourse Han/shaoshu minzu, nation/ethnicity, security/danger, and backwardness/modernity are mutually reinforcing dichotomies which reproduce minzu as something dangerous which is to be overcome by Zhonghua Minzu. In this discourse ethnicity becomes associated with the ‘negative’ side of the emerging dichotomies: backward, passive, and dangerous. It becomes understood as a threat from the outside, which must be secured against to produce a stable and prosperous nation.

Chinese anthropology and minzu theory on China, like the State Council quotes in section 1, frame China as a multi-ethnic nation. This is positioned in stark contrast to the “Western” nation-state and its historical experiences of nationalist conflict instead of tuanjie in China39. Outgoing President Hu Jintao used his final speech to the 18th National Congress of the CP to announce that “we will never copy a Western political system” (Xinhua, 2012e). This rejection of “the West” is not simply about institutional reform but about identity politics in which Chinese leaders orders the world into two civilisational camps separated by distinct cultural characteristics. This discourse is best understood as a performative enactment of internal/external boundaries: China is not the West and is unified not divided by nationalism. Minzu as marker of difference is thus subsumed into a larger national identity, which produces a

39 For example, see Pan (2008b); Zhang (2006), Ruan (2004), Fei (1980; 1988); Ma (2007).

85 unified, non-. To be minzu is to be Chinese and to be Chinese is to be non- Western. Minzu as a political category imposed from the outside is not the same as the English term ethnicity which includes self-identification. Of the more than 400 groups who applied to be a minzu, only 56 were eventually accepted (Fei, 1980, p.165). Fei Xiaotong’s (1980) review of the process of minzu shibie discussed in chapter 1 explained how researchers had to convince the chuanqing of Guizhou that were in fact Han and groups of self-identified Mongolians in Hulun Buir, that they were not Mongolian but Dawoerzu (Daur). This process was “scientific research work” which used historical records to determine the genealogical lineage and migrations of these peoples to tell them which minzu they belong to (Fei, 1980, p.170-174).

Chinese anthropologists today use the terms clan and tribe which had been eradicated from the social vocabulary by the party in 1962. Scholars such as Ma Rong (2007) and Pan Jiao (2008b) have for very different reasons suggested that China abandon the term minzu and instead use zuqun. This is partly because zuqun is a closer translation of the word ethnicity as it tends to include self-identification. For Ma Rong, “good ethnic relations” are necessary to reduce “operational costs” which hinders the rise of China as an economic power at the global level (Ma, 2007, p.3). Ma Rong sees self-identification with China as a means to international power and development which links the security of the nation to its identity. Ma argues that the construction of the minzu category has broken with Chinese tradition. He recommends the party-state now should “de-politicise” and “culturalise” ethnicity by abandoning minzu and allowing people to identify with their zuqun and their (political) national identity (Ma, 2007, p.2- 5). The Hua /Yi distinction thus should guide ethnic relations and “less civilised” groups can become “civilised” as they are attracted to Chinese civilisation in the same way as “barbarians” of the pre-modern era (Ma, 2007, p.5-7). Ma Rong’s (2007) argument is that conversion into Chinese civilisation is best understood as the development of shaoshu minzu. The security of the nation thus depends on the disappearance of non-Han identities through fusion into Zhonghua Minzu with the Han at the centre. The “tolerance of Chinese civilisation” is extended to those who accept acculturation into “Han-centred” Chinese civilisation and “discrimination” awaits those who do not accept acculturation (Ma, 2007, p.6-7). Ma is speaking in the language of the reformers such as Liang Qichao and their hope for “fusion” of the 5 races to unite a strong China. This challenges the party-state’s approach towards nation-building based on 56 recognised and institutionalised minzu. However, it does so from within this discourse. Ma argues that nation-building in China should assimilate minorities into Han-ness for the sake of development and national security. The language of modernisation and development is not supplanting earlier conceptualisations of difference but reconfiguring them so that Hua /Yi comes to mean developed/backward and Han/Shaoshu Minzu.

86

Ma Rong’s recommendations have stimulated considerable debate across Chinese academic circles. There has been some resistance to these ideas. Ruan Xihu (2004) argues it would contravene the constitution not to recognise minzu identity but zuqun would only split the nation by splitting the Han into separate zuqun. In 2012, the State Ethnic Affairs Commission set up a website for this “battle of opinions” which sets two contrasting views on ethnicity against each other; the “salad bowl” attributed to the Soviet Union and the “” exemplified by the US (Zhongguo Minzu Zongjiao Wang, 2012). That the “salad bowl” is framed as “1st generation” of policies and the “melting pot” as the “2nd generation” suggests changes are expected. The highly influential IR scholar, Hu Angang of Tsinghua University, wrote that a second generation of minzu policies are essential to transform “self-identification” (rentong) from minzu through “fusion” into a race-state (guozu) (Hu, 2012). This means following Ma Rong’s de-politicisation model by removing minzu from identity cards, abandoning the autonomy system, and the promotion of Mandarin Chinese-medium education (Hu, 2012). Even the party’s leading spokesman for ethnic affairs, Zhu Weiqun recommended these measures and wrote in the party journal Xuexi Shibao that ethnic inter-mixing and fusion are necessary for economic development and the “strengthening of the greater Chinese race” (Zhu, 2012). Wang Yang, the former secretary of the CCP in and architect of the “Guangdong Model” based on pro-market reform, suggested after the violence of July 2009 in Xinjiang that ethnic minorities policies need to be readjusted (Leibold, 2012a). Smith (2011) and Leibold (2012a) suggested this may point to progressive change in minzu policies. However, Wang’s unexplained and unexpected removal from office in the late 2012 leadership shuffle now suggests change is unlikely to be rapid or radical.

Rather than see these moves towards change as a second generation or a new phase of minzu policies, these are better understood as a reconfiguring of difference through earlier configurations. The vocabulary of race-state (guozu) is drawn from Republican era debates where Liang Qichao used the term to describe the highest level of human evolution. It was thus necessary for China to become a Guozu in order to survive in a “survival of the fittest” (Leibold, 2007, p.9-11 & 32). This social Darwinism was drawn from European racist discourse and given that Ma Rong’s appeal to the past through the Hua/Yi distinction as a mode of organising difference, there is little innovation here. Instead these debates reorganise earlier discourses to address contemporary concerns of “China’s rise” and “scientific development”. One overlooked perspective in Chinese anthropology which could contribute greatly to this debate is that of Pan Jiao of the Central University for Nationalities. Pan’s (2008b) position that the categories ethnicity (minzu and zuqun) and nation (guojia) are primarily distinguishable through power relations such that what counts as a nation or an ethnicity depends on the power the group holds. This relatively transparent debate is being conducted almost entirely by male Han Chinese politicians and academics within the boundaries of two options (minzu vs. guozu). This

87 illustrates that in China the power to articulate who we are lies largely in the hands of politically well-connected Han men. They are positioned relatively highly in Chinese power relations. These relations bound the possibilities of identity articulation within the binary options of the status quo or assimilation into guozu with the Han at the centre. These debates which articulate China and “the West” through a seemingly niche study of ethnicity in contemporary China are becoming increasingly important to understand. China’s power to articulate identity politics at the global level is only to increases with its material power.

China is culturally spatialised through dichotomies which enforce a hierarchical regime of belonging and difference at the same time. For example, in the 2012 leadership struggle the “Chongqing model” of Bo Xilai has been de-legitimised for reinstituting political education and “sing red songs” by state-media, bloggers, academics, and even the World Bank40. At the same time, the party-state has made minzu tuanjie education compulsory across all state institutions in Xinjiang to “promote minzu tuanjie consciousness…with patriotism as its core content” (XUAR Government, 2009, article 3). The fact that formal “political education” is a near-universal experience in Xinjiang but deemed out of date for the rest of China only reinforces the uneasy incorporation of Xinjiang into China. The texts used in Xinjiang as minzu tuanjie education for schoolchildren, university students, and cadres introduce some concepts which are rarely encountered in other regions of China, such as “ethnic extinction” (minzu xiaowang). Minzu tuanjie education also introduces ethnicised ways of framing other political concepts which are frequently heard in all regions of China, such as “scientific development” (kexue fazhan). Within this narrative of Xinjiang history, the unbroken “path” of Xinjiang culture’s “historical evolution” runs from the pre-stone age to the present day. It runs towards cultural “fusion” and “” of the modern “harmonious” culture of the Chinese nation under the leadership of the party-state: “Xinjiang’s culture has since ancient times always been based on fusion…this is the historical development of the process of evolution and the unceasing raising of Xinjiang culture” (Ethnic Affairs Commission, 2009, p.16-17). Particular importance is thus stressed on cultural evolution in Xinjiang to write the history, the present, and the future of the region as one of “fusion” of Hua and Yi and of different minzu.

General Zhang Zhizhong, who negotiated the dismantling of the East Turkestan Republic, followed ’s ethnotaxonomy of Xinjiang into 14 indigenous ethnic groups. In many ways, this ran counter to the Stalinism of the emerging minzu project (Jacobs, 2008, p.551; Millward, 2007, p.208). Today, these 13 minzu are still officially recognised by the party-state as the “long-term residents” of Xinjiang (Ministry of Information, 2009, p.50; Ethnic Affairs Commission, 2009, p.17). However, the texts used for minzu tuanjie education make no

40 For example, see: Xinhua (2012), Inside Out China (2012), Freeman and Wen (2012), and World Bank (2012) respectively.

88 mention of multiculturalism except, as we shall see below, with regard to examples of Han migration. The 50 Whys reiterates this Han-centric discourse in its framings of Xinjiang as “backward” until “liberation”. However, it also does so within its references to “multi-culturalism since ancient times”41. These say nothing of the other 12 minzu but remind readers “the history of the Han in Xinjiang is very early – from before the time of Zhang Qian42, there have always been Han living in Xinjiang” (Ministry of Information 2009, p.50). This omits the logically essential contribution of the multiple non-Han ethnic groups to multiculturalism to emphasise the importance and presence of the Han.

In these texts, “multiculturalism” refers not to multiple ethnic groups within a pluralistic political framework but to the demographic transformation of Xinjiang where Han become the majority. The “active spirit” of the Han and the Chinese state is said to have driven cultural “fusion” and the development of Xinjiang throughout history with the “support” and “simple, uncomplicated” assistance of the Shaoshu Minzu armies (Ethnic Affairs Commission, 2009, p.87). The unbroken lineage of Han-ness drives this fusion and this runs from the Han and Tang dynasty armies, to Qing era land reclamation, to the entrance of the bingtuan, and finally to the reform-era mass incursion of Han migrants Ethnic Affairs Commission, 2009, p.13). Han migration makes Xinjiang a “multicultural place” and timelessly so such that “all groups who live in Xinjiang today emerge from successive migrations” (State Council, 2009b, p.31). The “frontier culture” of the Han is represented as the active driving force of multiculturalism and fusion since ancient times where Shaoshu Minzu are relationally defined as its passive, backward recipients. Han-ness is in many ways represented as non-ethnic and normalised as the nucleus of the nation which is both multi-cultural and Han-centred. Instead of being framed as an ethnic category, Han-ness becomes objectivised as a force for development and the standard of the nation. The Han simply cease to be “ethnic” in this formulation because they are represented as beyond minzu. In these texts, Shaoshu Minzu (ethnic minority) is interchangeable with Minzu (ethnic group) as ethnic-ness becomes equated with non-Han-ness and defined as an absence in opposition to the presence of Han-ness43. This absence is then something to be overcome through fusion and nation-building. It is here that the concept of “ethnic extinction” (minzu xiaowang) emerges in these texts44.

41 That Xinjiang is “multi-ethnic” (duo minzu) or “multi-cultural” (duoyuan wenhua) is a common theme in these texts. For example, see State Council (2009b, p.3). “The Three Evils” then “ignore” the multi-cultural history of Xinjiang by supporting Uyghur separatism (Ministry of Information, 2009, p.50). 42 Zhang Qian was a 2nd century BCE Chinese envoy who visited Xinjiang and played a key role in establishing silk road trade. 43 The text, Ethnic Minority Policy, if literally translated would be Ethnic Policy (minzu zhengce) but it is an elaboration of policy in regions populated by Shaoshu Minzu. 44 It ought to be noted that this concept appears in the study texts for ‘internal circulation’ only such as The 50 Whys and Common Knowledge. There is no official explanation as to why this is the case. However, it may reflect an official concern regarding the sensibilities of an international audience. Representing China as attractive to foreign capital and tourism, as well as worthy of cultural recognition is a huge concern of the CCP. This has been framed by some scholars as “soft power” (Nye, 2004) or China’s

89

The party-state’s ethnic unity education materials in Xinjiang state its explicit goals as “ethnic extinction” and the “fusion” of 56 ethnic groups into the common identity of Zhonghua Minzu. Its contention that “only if ethnicity exists can there be an ethnic problem” suggests that the party- state’s approach is the eradication of ethnicity over the long term (Ethnic Unity Education Board, 2009, p.37). This phrase is drawn from the identity politics of the (Shijian Bianji Bu, 1965, p.220), a period of open hostility to all identities deemed “traditional” and ethnic. Looking at these identity politics on the ‘periphery’ challenges the orthodox understanding of China as in gradual transition from “communism” and “authoritarianism” to the “free-market” and “liberalisation”. Here we see the inscription of difference under pluralism but without the liberal tolerance for different groups or individuals to articulate different ends:

‘Ethnic extinction is an inevitable result of ethnic self-development and self- improvement……It is the final result of ethnic development at its highest stage……in this big ethnic family every ethnic group has a higher level of identification – Zhonghua Minzu’. (Ethnic Unity Education Board 2009, p.17 & 79)

Ethnic extinction is thus understood as an inevitable step in the teleological progress where the economic development of minzu will compel them to move up the ladder of cultural evolution and lose their consciousness as groups. The party-state’s fused ideology of Marxist historical materialism and Social Darwinist cultural evolution understands history as a progressive flow towards national community and modernisation. Modernity then entails the disappearance of ethnic identification and its symbols. The stated end-point of this teleology is a global classless society where all nations disappear as Karl Marx and Mao Zedong envisaged. However, “ethnic extinction” is dependent on the continuation of the economic development experienced through the reform period under “market socialism” (Ethnic Unity Education Board, 2009, p.17). National prosperity and the disappearance of ethnicity it entails require temporary ethnic unity such that peoples in Xinjiang understand themselves as members of a Chinese national community. “Ethnic minority work” (minzu gongzuo), including minzu tuanjie education and Mandarin-medium education with the “scientific development outlook as their basis” are essential to the “common unity and development of every ethnic group” (XUAR Dept of Information, 2009, p.86). This ideological work then is a temporary measure to eliminate the leftover remnants of ethnic identification from earlier historical stages of development and transform ethnic identification into national identity (minzu tuanjie).

These materials claim the constitutional right to “maintain and develop” minority languages remains unaffected in this model of “multiculturalism”. However, the reduction of the number of

“charm offensive” (Kurlantzick, 2007). One can assume the notion of “ethnic extinction” would be internationally challenged by liberal theorists of multiculturalism and human rights organisations.

90 languages in the world and the expansion of Mandarin Chinese use is said to be an “inevitable outcome” of “modernisation” (Ministry of Information, 2009b, p.94-95). The modernisation of values and “ways of thinking” (siwei fangshi) in Xinjiang are represented as an “historical inevitability” (Ethnic Affairs Commission, 2009, p.23). This orders self-identifications such that the category Shaoshu Minzu and the associated symbols of language are obstacles to be overcome by a unified Chinese national community. Chapter 5 will discuss how minzu tuanjie is performed to mass audiences in ways which not only place Han at the centre but objectivise Han-ness as Chinese-ness. Minzu Xiaowang is part of this discourse of conversion. The concept is not applied to the Han and Han-ification (Hanhua) is promoted in ways which conceals the minzu-ness of the Han category. Han-ification is promoted as an inevitable outcome of modernisation such that unproblematised Han-ness becomes the trans-ethnic identity of an assimilative race-state.

The official writing of the history and future of the Chinese nation is, as Duara (1995) argued, an attempt to secure unity through time for the contested and contingent category of China. However, it fails to secure this unity because its central motif is difference and internal boundaries. Where it offers unity this entails conversion rather than tolerance of difference. In this nation-building narrative, people are either on the side of historical progress and modernisation or left behind. The Three Evils’ (san gu shili) opposition to the “sacred mission” of national unification is said to run “against the flow of history” and “modernisation” thus it will “inevitably lose” (Ethnic Unity Education Board 2009, p.60). Resistance to the party-state’s nation-building then is framed not simply as evil but as futile. The 50 Whys frequently reminds readers that certain ethnic affinities such as wanting one’s ethnic kin to economically succeed are “reasonable” but that national identity must be predominant in any individual’s self identification (Ministry of Information, 2009, p.35-37). This is because Chinese-ness understood through discourses of Han-ness is posited to be the highest level of identification.

Ethnic identities are being criminalised through the securitisation and essentialisaton of the boundaries of Chinese-ness. The 50 Whys text tells readers are that only “separatists” and “terrorists” could possibly understand themselves in any other way, such as through reference to the categories of “Turk” or “Muslim” (Ministry of Information, 2009, p.47-52 & 99). These sections of The 50 Whys are directed at Uyghurs because their Turkic and Islamic self- identifications are represented as security threats to China 45 . Ultimately this discourse articulates identity to Uyghurs and frames alternatives as threats. It encloses their self- identifications into the Chinese sphere and securitises the boundaries of how they understand the relationship between their ethnicity and their nation. The party-state’s discourse on the

45 This will be discussed more fully in chapter 3.

91 security threat of ‘The Three Evils’ reveals less concern about armed organisations and more about people’s self-identification and interpretation of history:

The Three Evils believe that under the rule of the Qing Xinjiang became a Chinese colony. This ignores facts and distorts history. After the establishment of the Western Regions during the Han dynasty, every ethnic group identified with the rule of the central government. (Ministry of Information, 2009b, p.47-50)

This exclusion of alternative understandings of history as ‘evil’ illuminates how the dichotomies of modern/backward, active/passive, and security/danger reinforce each other and are linked to an articulation of the inside/outside of Chinese-ness. This articulation of nationhood produces an inside which is modern, secure, and defined through Zhonghua Minzu. China is relationally defined against the purported backwardness, danger, and ethnic-ness of the ‘outside’, which is physically inside China. The boundaries of what it means to be Chinese and what it means to be Chinese-Uyghur are demarcated such that everything outside the party-state’s narrative is a form of separatism, terrorism, and extremism. It is claimed that “mistaken” ethnic identities, such as those defined by reference to minority languages, only exist due to this dangerous influence of the “inside-outside” threat of The Three Evils (Ministry of Information 2009b, p.55- 59). The party-state is constructing inside-ness by reference to what Winichakul called the “constitutive outside” and Judith Butler, the “negative elaboration” of the Self. However, this outside is inside the borders of China in the form of the Uyghur internal Other which threatens to disrupt the Self by thinking of themselves as Turkic Muslims. Uyghurs who interpret the arrival of Qing armies as invasion rather than liberation, identify themselves as Turks, or argue for the promotion of education using Uyghur as the medium of instruction are excluded from the national community. They are being linked to a threatening past of backwardness and a dangerous outside of security threats. These mutually reinforcing dichotomies symbolically construct national community by imbuing historical representation, self-identification, and language use with normative meanings positioned in a social hierarchy. Nation-building and modernisation are conversion projects to convince Uyghurs and Han that they are ‘inside’ the same timeless and progressive but hierarchically structured community. This teleology includes all ethnic groups but excludes those who understand themselves in different ways as counter to its progress.

The oft-seen Deng Xiaoping slogan in these texts and on the streets of Ürümchi, “stability overpowers everything” (wending yadao yiqie), gives the impression that nation-building in Xinjiang is all about stability. Maintaining the stable rule of the party-state is indeed a top priority. However, the goal of nation-building in Xinjiang is not simply about political stability but social transformation. Nation-building promotes the transformation and conversion of identity so that Han and Uyghurs understand themselves as unequal members of the same

92 national community. It is best to understand nation-building within the dichotomies illuminated here, which are intertextually linked to the transformative concepts of “fusion” and “ethnic extinction”. Nation-building seeks to convert shaoshu minzu so that that they identify with the Chinese nation first and their ethnicity second. The political system of regional autonomy and its stability is officially represented as the means to this transformation of self-identification. After coming to power, regional autonomy was adopted by the CCP as a means to placate calls for independence, particularly on the former Qing Frontier (Clarke, 2007, p.278-284). The system is that of a highly limited territorial-based cultural autonomy for regions with high proportions of non-Han 46 (Clarke, 2007, p.278). Official explanations for the regional autonomy system for minority regions today begin with the now familiar dichotomising representations of Xinjiang that prior to “liberation” Shaoshu Minzu were “enslaved”, “oppressed”, and their economy was “backward”. Since the “liberation” of Shaoshu Minzu, regional autonomy is said to have made them their “own masters”; they enjoy equality, freedom, and development, most notably “tall buildings” (Ministry for Education, 2008, P.72-73). However, unlike the non-autonomous provinces, autonomous regions are in practice directly answerable to and under the direct supervision of the State Council, the highest level of central government. This ensures a level of centralised control over policy and its implementation not seen in the non-autonomous areas of the PRC.

Articles 9, 14, 55, and 71 of the National Law on Regional Autonomy (2001) state that the purpose of the autonomy system for ethnic minority regions is to ensure that ethnic minority regions can never be separated from the motherland and are “modernised” by raising their “socialist consciousness” and their “scientific level” (kexue wenhua shuiping). A stable “autonomy” system is thus part of the nation-building discourse of identity-security. This produces both the territorial boundaries of the PRC and Xinjiang’s position as an Orientalised frontier within China. The division of Xinjiang into administrative units after 1949 was in many ways a form of gerrymandering. For example, to counter Uyghur autonomy, Daur and were given Tacheng and Emin autonomous counties despite only having 17% and 12% of the populations respectively (Bovingdon, 2004, p.13-14). The establishment of these “sub- ” served the major goals of entrenching the idea that Xinjiang belonged to the 13 officially recognised ethnic groups and of countering the demographic and potential political weight of the Uyghur (McMillen, 1981, p.66–70; Clarke, 2007b, p.333). However, the autonomy system like the renegotiation of the meaning of the minzu concept is an example of where the discourses of nation-builders are turned against themselves. The establishment of the Xinjiang Uyghur Autonomous Region in 1955 has provided Uyghurs with a more distinct and bounded “territorial frame” for their grievances and political aspirations (Bovingdon, 2004, p.3). This will

46 Namely the Xinjiang Uyghur Autonomous Region, Inner Mongolia Autonomous Region, , Hui Autonomous Region, and the Zhuang Autonomous Region.

93 be discussed more fully in the next chapter but this is a useful reminder that nation-builders have no monopoly over nation-building. The model of inclusion offered by the party-state and in many ways by anthropologists is inherently exclusionary of alternative modes of self- identification and historical analysis. As we will see in Chapters 6 and 7, inclusion will continue to be understood as assimilation by non-Han if it does not find more attractive ways to include non-Han in its formulation. This means promoting a “new generation” of minzu policies which do not place ascriptive and inconvertible concepts of race and lineage at their centre.

Conclusions

How history is written and the social categories it employs are framed are central to the Chinese party-state’s nation-building project in Xinjiang. Like the construction of European nations, the creation of a “long and proud national historical consciousness” which projects the modern spatial domains onto ancient history is the primary discursive step of this project. Official histories of the party-state in Xinjiang imagine the boundaries of self/other as a series of sovereign and bounded national entities. Section 1 showed how this particular imagined community relies on the representation of Xinjiang as an indivisible component of the Chinese nation. Section 2 showed how this discourse also relies on spatial divisions between regions of China (neidi vs. bianjiang) and between ethnic groups (Han vs. Shaoshu Minzu). These internal boundaries were exemplified in the idea of an unbroken ethnic Han lineage, occupying and developing the territory since the Han dynasty through to the contemporary party-state. These official texts on ‘national’ history produce representations of social reality which individual humans mediate in order to comprehend reality as well as to succeed in education or politics in Xinjiang. These representations are based on oppositional and mutually reinforcing framings of self/other, inside/outside, backwardness/modernity, and security/danger. Section three showed how these representations informed a discourse on “ethnic extinction” where ethnic minority policies are directed towards producing self-identification with China and the elimination of non- Han ethnic identities. This division between Han and non-Han form the basis of the text of nation-building. This dichotomy produces a discourse of what it means to be Chinese and what it means to be Shaoshu Minzu through the “constitutive outside” and a “negative elaboration” of the Self. The former dominate where the latter must be dominated.

The official historical narrative projects the past onto the present and onto the future such that the dichotomies of self/other, inside/outside, and security/danger are represented as timeless and universal questions of historical national foundation and contemporary security problems. The historical division between pre and post-“liberation” Xinjiang as backward and modern respectively articulates nationhood in ways which reinforce the ethnic boundaries and difference

94 between the categories of Han and Shaoshu Minzu. Han-ness and Minzu-ness only have meaning in relation to one another and through the mutually reinforcing discourses of backwardness/modernity, frontier/neidi, and active/passive cultures within this nation-building discourse. By projecting a backward Otherness onto the categories Turk and Muslim the party- state securitises identity. It seeks to transform how Uyghurs understand themselves but also to exclude those Uyghurs who continue to identify themselves as Turks and or Muslims. This narrative does not reflect an objective reality of ‘China’ ‘out there’. It is a discourse, which through repetition produces the capacity to make meaning of reality and categorise persons into bounded minzu within bounded nation-states. The timeless self/other which these writings produce is not simply about external boundaries but about ordering the nation around internal boundaries of Hua /Yi but reconfigured in the vocabulary of modernity and development.

Rogers Brubaker’s described the juxtaposition of ‘good’ against ‘bad’ ethnic nationalism as a “Manichean myth” (Brubaker, 1999, p.62). They are not mutually exclusive categories. Citizenship, based on the multi-ethnic formulations of nations, is just as inclusive/exclusive and has as much a “cultural component” as ethnic formulations of nationhood (Brubaker, 1999, p.61). The difference is that one consciously foregrounds these cultural components where the other conceals them. The party-state’s official account of history produces a form of national identification which includes all 56 minzu but excludes their perspectives if they understand themselves or their history in ways which do not reflect the supposedly ‘good’ ‘civic’ nationalism of minzu tuanjie. Chinese nation-building shows us how states can conceal and objectivise cultural identities under the rubric of civic nationalism. The teleological theory of cultural evolution in this narrative frames ethnicity in opposition to nationhood and as something which will be overcome through cultural “fusion” and modernisation. However, “fusion” is an enclosed, nationalised end-state within Zhonghua Minzu rather than an ongoing, negotiable process in an increasingly interconnected and globalised world. This enclosure conceals, obscures, and threatens the ethnic, linguistic, and religious links Xinjiang has with Central Asia and limits the discursive space available to people to articulate their own identities. The direction of the cultural flows driving fusion in Xinjiang is strictly “inward” and never “outward”. Xinjiang then becomes represented as an empty, passive space to be acted upon rather than an active participant in national affairs let alone world history. However, despite this “fusion” narrative the Han continue to be represented as an unbroken lineage while other groups remain “ethnic”. Uyghurs are being included in the timeless Chinese nation but into a hierarchically ordered social position behind the Han.

Zhonghua Minzu is represented in Xinjiang as a unified, multi-ethnic, and developing community. This representation is relationally produced against its mutually constitutive textual opposition of the separatist, ethno-centric, and backwardness offered by the Uyghur past. These

95 dichotomies reproduce a discourse of China as a culturally defined nation and as an ethnic hierarchy. This hierarchy involves securitising a particular way of being Uyghur and being Chinese such that nationhood is a modern community and ethnicity is a backward threat. In the name of security, minzu thus must be overcome. The party-state promotes Han-ness as the active force driving this teleological progression. However, this means that China is performed through Han-ness where “backward” minzu are to be absorbed and converted or excluded. Han chauvinism did not disappear with a class-based revolution nor did it disappear with the shift to a focus on development. It has merely been reconfigured so that Han-ness is the scientific progression into which other minzu are to be assimilated. This chapter explored how the official narrative of the Chinese nation in Xinjiang is performed through education in Xinjiang. The next chapter will explore how this narrative of Chinese-ness is securitised through the exclusion of alternative forms of Uyghur-ness in the official discourses of East Turkestan and “The Three Evils”.

96

Chapter 3: East Turkestan and the Production of the Internal/External Boundaries of Chinese-ness

Introduction

Chapter 2 explored how the is written in ways which articulate a teleological unfolding of and cultural evolution towards Han-centric Chinese-ness. This chapter is about the how these discourses on history inform the exclusion of Turkic-ness as culturally ‘outside’ China and the relational securitisation of Han-centric Chinese-ness in contemporary official texts. The analysis will explore how Chinese-ness is performed and securitised through the exclusion of forms of Uyghur-ness. The boundaries of community are neither solely internal (‘cultural’) nor external (‘territorial’) in this analysis of national boundaries. The identification of both insiders and outsiders is central to the production of the boundaries of any purported sense of ‘we’-ness. Community implies a feeling of inclusion based on common practices amongst members but this demands exclusion because it distinguishes itself from members of other communities (Cohen, 1985, p.14). Exclusion and inclusion are thus intertwined aspects of the same articulations of identity. This chapter will shift focus to how the inclusionary nature of nation- building logically demands exclusion of characteristics deemed unbefitting of the nation’s essence. By including some Uyghurs as members of the hierarchical Zhonghua Minzu, what characteristics and what types of Uyghur-ness does this exclude?47 The approach here as in previous chapters is less about uncovering previously hidden “facts” but to examine how the internal logic of discourses on Xinjiang’s identity politics produces categories of self- identification and how they can be taken to be objectively real by states and people. Following Said (2003), Shapiro (1998; 1999), Weber (1998), Butler (1999), and Campbell (1998) the purpose here is to explore how these discourses offer the capacity for humans to make meaning and how they produce the boundaries of truth/false and security/danger.

Communities tend to respond assertively to encroachment or transgression of the boundaries which encapsulate their identification because they feel their identity is under threat (Cohen, 1985, p.91). In international politics these threats are codified by states through “discourses of danger”, which demarcate the boundaries between safe insiders as the referents of security and dangerous outsiders as threats (Campbell, 1998, p.1-3). In studying Xinjiang we will see how these boundaries are securitised in ways which collapse and fuse the distinctions between

47 The focus in this chapter is on the exclusion of characteristics which are articulated as “ethnic”. This does not imply that there are not other exclusions based on gender, sexuality, or age in articulations of Chinese-ness. However, the focus on ethnicity emerges within these texts which relate ethnicity and Islam to security.

97 domestic and international politics and between internal and external boundaries. Securitisation is rethought here as a performative enactment in the ongoing production of we-ness. During the Republican era (1912-1949), CCP discourses on Chinese nation-building drew considerable attention to these questions of inclusion and exclusion. Mao Zedong asked: “Who are our enemies? Who are our friends? This is a question of the first importance for the revolution. The basic reason why all previous revolutionary struggles in China achieved so little was their failure to unite with real friends in order to attack real enemies” (Mao, 1926). Politics in China has changed dramatically since 1926 with Nationalist, then Communist, and now state-capitalist48 regimes ruling China. However, the question of who are the friends/enemies of China in the reform era remains central to party-state discourses on identity as it grapples with its growing international power. Resistance to being included in the official discourse of Zhonghua Minzu as a self-identifying nation of 56 united minzu presents problems of identity-security. These questions of identity politics are more acute on China’s frontiers because they continue to internally challenge what it means to be Chinese by resisting their interpellations as Chinese.

The first section will show how the official discourse of East Turkestan designates Turkic-ness as an outside danger relationally producing the secure Han ‘inside’ of China. This discourse produces internal/external boundaries, whereby the external territorial borders and internal ethnic boundaries are projected onto and mutually reinforce one another. These discourses are central to Chinese nation-building in Xinjiang to such an extent that the identification of ethnic enemies within as a “constitutive outside” is crucial to producing a unified national Chinese self. Section Two will analyse how the party-state explains incidents of Uyghur unrest and dissent in Xinjiang through the discourse of “the inside/outside Three Evils” (jingneiwai sangu shili)49 of “terrorism, separatism, and extremism”. It will show how shifts in representations of security and Uyghur-ness have been contingent on changing political contexts. This section will explore how the case of Xinjiang in China can help us understand how states simultaneously perform the theoretically indivisible internal/external boundaries of community inclusion/exclusion. We will see how state-centric security discourses, like all discourses on community boundaries, tell people who they must not be (Turkic Muslims) and who they must become (Zhonghua Minzu).

The third section will explore how the indivisibility of the internal and external boundary is applied to the party-state’s ethno-spatial boundary drawing within Xinjiang. This focuses on how boundaries are drawn between more and less secure and more and less modern regions linked to the ‘outside’. This spatial boundary drawing is a taken for granted part of domestic

48 This is described by the party-state as “market-socialism” because markets are ostensibly utilised in China not for the benefit of individual interests but society as a whole. For example, see: Ministry of Information, Theoretical Department (2009a). 49 “Sangu Shili” can be literally translated as “three forces”. However, the party-state’s official translation used in English language White Papers is “The Three Evils”. For example, see State Council (2009a; 2009b).

98

Chinese politics and the party-state seeks to avoid “internationalisation” of the “Xinjiang problem” and minzu questions (Ministry of Information, 2009, p.8). However, it is frequently linked to both a constitutive outside but also a physical outside through discourses of the threat of terrorism linked to Islam and “the West”. The party-state performs the domestic and the international in ways which make the internal and the external indivisible. Despite the party- state’s attempt to demarcate distinct identity boundaries based on national borders, this attribution of outside-ness to those physically inside China blurs the lines between the internal and the external and between the domestic and the international. Xinjiang is thus framed as an internal/external security dilemma for the Chinese state.

3.1: East Turkestan and the Securitisation of Internal/External Boundaries

The earliest known use of the term Turkestan comes from 9th century Persian sources to denote the lands of Transoxiana and contemporary Xinjiang (Millward, 2007, p.55-56). The suffix ‘stan’ means ‘place of’ or ‘land of’ in Persian. Turkestan then translates as ‘land of the Turks’ (Dwyer, 2005, p.52). Ildikó Bellér-Hann’s Community Matters (2008) analysed indigenous sources written in the East Turki dialect from Xinjiang in the period 1880-1949. These sources, including local historical accounts and essays collected by missionaries, used the ethnonym Turk or Turki almost universally (Bellér-Hann, 2008, p.27-29 & 51). The term Turkestan came to prominence amongst Turkic nationalists across Central Asia in the 19th century. Turkic nationalists sought to build a nation, which included the establishment of educational institutions bearing the name such as the Turkistani Youth Turk Union founded in Istanbul 1927 promoting the use of (Dwyer, 2005, p. 51-52; Schluessel, 2009, p.388). The terms East Turkestan and West Turkestan emerged in the works of Russian scholars in 1829 to divide Turkestan into eastern and western sections along the boundary of former Russian control at the , which today separate Xinjiang from and Kyrgyzstan (Kamalov, 2007, p.34). Turk does not distinguish between speakers of different Turkic languages (Bellér-Hann, 2008, p.51). This frames belonging through language. It is as far removed from the contemporary map of national borders and top-down ethnic classification projects as is any objectively existing ‘China’ based on 5,000 years of Hua culture. This section will show how official party-state discourse on East Turkestan is a crucial part of nation-building through the interlinked processes of inclusion and exclusion in Xinjiang. These discourses demarcate the boundaries of Chinese-ness by designating Turkic-ness as outside these boundaries and as a danger to the maintenance of identity-security.

When I asked Uyghur interviewees “what was Xinjiang called before the Chinese arrived?”, one typical response came from a young male student, who told me “this is East Turkestan”. While

99 the party-state seeks to delink the category of Uyghur from the broader category of Turk, this category remains a powerful symbol amongst Uyghurs to contrast themselves with the category of Chinese. It is de facto illegal for Uyghurs to even utter the term East Turkestan in the PRC today (Dwyer, 2005, p.52). For example, one Uyghur interviewee would only refer to the “old name for Xinjiang” and refused to specify the name because in his own words “I have to work for the government”. The fact that we were friends talking privately in the Pamir Mountains with no people for miles indicates the fear of expressing open identification with this history amongst Uyghurs. The term East Turkestan is outlawed as antithetical to national unity because it emphasises the Turkic ethnic identity and Uyghur ownership of Xinjiang.

Several Uyghur intellectual informants explained to me that the use of Xinjiang instead of Turkestan was only to avoid “offending” Han. Uyghur nationalist movements refute the name Xinjiang because of its colonial undertones (“new frontier”) while scholars of Xinjiang tend to use the name to avoid political controversy (Roberts, 2012, p.3). Today, Uyghur diaspora campaigns for autonomy and human rights in Xinjiang such as the East Turkestan Information Center (ETIC) and the World Uyghur Congress (WUC) continue to use the name East Turkestan to refer to Xinjiang. This usage was found throughout many interviews with Uyghurs in the region and represents a form of resistance to the party-state’s discourse on Zhonghua Minzu and “The Three Evils”. Nation-building in Xinjiang is seeking to impose a self-identification (Zhonghua Minzu) where one already exists (Turkestan) and in ways which seek to transform how people conceptualise community (state-centric nationalism vs. cultural identification). This ontology of human communities and their history is self-defined through language, ethnicity, and religion. However, this is anathema to the official Chinese understanding of the world through nationalism discussed in chapter 2 as a “world of minzu”, where 3,000 officially designated minzu are organised into more than 200 bounded, sovereign states (State Council, 2009a, p.1). This raises serious problems for the official Chinese model of nation-building because it is not only building a new community but a new type of community.

The Chinese party-state claims that Turkestan is not only an “undefined”, “geographical” term but a “political concept first put forward by old colonialists with the aim of dismembering China” (State Council, 2002; State Council, 2003) 50. The party-state frames all claims to Xinjiang’s historical separateness from the Chinese cultural sphere (zuguo) or the Chinese state (guojia) as by-products of nineteenth century Western imperialist’s “concoction” of Turkestan as a political unit to divide Chinese territory and “split the motherland” (State Council, 2002; State Council, 2003). According to the party’s historiography, after the British and Russian empires established Consulates in Kashgar in the midst of the so-called ‘Great Game’, they used their

50 The term Turkestan is still used by historians who work on history outside of China. For example, see Davies (1997) Europe: A History.

100 foothold in the north-west to foment discontent amongst ethnic minorities, creating the “illusion” that Xinjiang was separate from China (Li, 2007; State Council, 2002). In this narrative, the term East Turkestan is attributed to Western imperialists and Uyghur “terrorists”51 due to the association with the two independent Eastern Turkestan Republics established in Kashgar 1932-1933 and Ghulja/Yining 1944-1949 (Millward, 2007, p.ix). The 50 Whys repeats these discourses for a broad audience in Xinjiang telling readers “Turkestan is a reactionary concept”, which “since the 20th century is only used by an extremely small minority of separatists influenced by religious extremism and ethnic chauvinism on the international stage” thus following the “methods of old imperialism” in “fabricating a system of thought of so-called ‘East Turkestan independence’ since ancient times” (Ministry of Information, 2009, p.53-54).

The threat of imperialism from the ‘outside’ linked to dangers “inside” the Chinese nation has played a central role in modern Chinese history and Chinese politics. Nationalism and socialism were in many ways inseparable in the CCP’s discourses on Chinese nation-building prior to and after 1949 as Mao Zedong proclaimed the corrupt “foreign ministries” of the GMD as the “counting houses of our foreign masters” (Mao, 1923). Mao’s nation-building discourses during this period highlighted the threat of feudalism controlled by the bourgeoisie linked to the imperialist threat to the wellbeing of the Chinese people and the strength of the nation (Mao, 1932). Just as it was impossible to separate nationalism and socialism, it was also impossible to separate the imperialists outside China and the capitalist collaborators inside China. Imperialists and capitalists were securitised together as an inseparable threat to what it meant to be Chinese. The friends and the enemies of the revolution were both inside and outside China and were threats to the identity-security of China and not simply to the party.

The official Chinese narrative here which positions imperialism as ‘outside’ and specifically un- Chinese has been updated for a contemporary audience by mainstream Chinese civic intellectuals such as Zhang Weiwei (2012) and Hu Angang (2012b). Zhang and Hu both assert that imperialism and “hegemonism” are “Western” approaches to international relations which China will never pursue. However, these narratives all conceal China’s imperial history on its frontiers. Framing imperialism as un-Chinese overlooks the findings of Peter Perdue (2005) and James Millward (2007), who show that Xinjiang, as a modern administrative unit, emerges at the confluence of Qing, Russian, and British empires. These narratives then position those Uyghurs who identify with East Turkestan instead of China as un-Chinese because their identities are officially articulated as a result of manipulation from outside China. The party- state’s theory of cultural evolution discussed in Chapter 2 is applied to the idea of East Turkestan in its minzu tuanjie texts. These texts claim “East Turkestan” and “The Three Evils” are “remnants of class exploitation” from imperialists and “historical leftovers of old thinking”

51 For just a few examples of this account, see State Council (2002; 2003; 2009b).

101 which “threaten national security” (XUAR Information Dept, 2009, p.84-85). East Turkestan is then a group of “separatists” who seek to “unite all the ethnic groups speaking the Turkic language” (State Council, 2002). East Turkestan is thus framed as an outdated form of identity- politics based on language and religion. It is also represented as a dangerous mode of self- identification to be transcended through absorption into the Chinese nation.

Many Chinese scholars in Xinjiang reproduce this discourse for academic and popular audiences. These include the Xinjiang University historian Pan Zhiping, awarded for contributions to the party and social stability in Xinjiang as part of the “theoretical vanguard” discussed in chapter 2. After the violence of July 2009, Pan Zhiping’s book, The History and Present Situation of “East Turkestan” (2008), was placed in the “hot topics” section of Xinhua bookstores all over Xinjiang as well as in Xinhua and All Sages bookshops in Beijing. Pan claims the idea of “fake” “Turkic history” is based on “, legends, and folk-tales”, which instigates “crazed minzu separatist activities” (Pan, 2008, p.89). He argues that prior to the “rise of capitalism in the West” and the penetration of China by “imperialists”, Xinjiang had always been “united” (Pan, 2008, p.24). The Pan- and Pan-Turkism of “East Turkestan” are then framed as “responses to “Western politics” (Pan, 2008, p.24 & 28). “East Turkestan” is a “by-product of 19th century geo-politics” which introduced the concept to Xinjiang through traders from Russian “Tartarstan” and the influence of Turkish politics on students from Xinjiang (Pan, 2008, p.72-74). The Turkic threat originates from imperialists and capitalists ‘outside’ of China seeking to convince Uyghurs they are Turks for geo-political ends. This seeks to silence and erase the category of Turk from history and from the present. Furthermore, the idea that disunity within the territory of China emerges only with 19th century intervention of the antagonistic politics of ‘the West’ reinforces the official relational framings of ‘China’ and ‘the West’ in chapter 2. ‘China’ is the multi-ethnic, harmonious, and collectivist mirror-image of the nationalist, exclusionary, and individualistic ‘West’.

Pan Zhiping argues that “East Turkestan” remains a “reactionary” threat to “national security, stability, and survival” (Pan, 2008, p.73). Pan, like the party-state is framing identification with East Turkestan as a threat to Chinese identity-security and the survival of Zhonghua Minzu. Belonging to East Turkestan through pre-modern cultural boundaries based on the Turkic language national borders is thus framed as a move backwards in terms of cultural evolution. Nevertheless, this ‘civic’ model conceals the cultural boundaries of contemporary China, which demarcate who is Chinese through ideas about pre-modern Chinese civilisation, such as Mandarin Chinese abilities understood through the Hua /Yi discourse. Like the party-state’s texts discussed in Chapter 2, for Pan, “East Turkestan” and “minzu separatism” are not only evil but futile as they are “completely defeated histories” (Pan, 2008, p.146). Turkestan as a form of identification is thus being dismissed as “fake” but it is also criminalised as an internal/external

102 threat to the ‘real’ China. The threat of this identification crosses national borders into Islamic, Turkic speaking Central Asia and the capitalist “West”. This relationally performs the inside of China by securitising its boundaries to exclude East Turkestan and ‘the West’ as a dangerous ‘outside’. The production of Turkestan as a threat (exclusion) is a nation-building practice of relational articulation of what it means to be Chinese (inclusion).

Arienne Dwyer (2005) reminds us that “Language policy crucially involves not only codifying words themselves, but also codifying their use. This has been accomplished through the repetition of key words or phrases in the public discourse” (Dwyer, 2005, p.50). The ETIC and the WUC are designated as international “terrorist” organisations by the party-state, partly for their use of the name East Turkestan (Ministry of Information, 2009, p.17-19). The generic term for East Turkestan (dongtu) appears frequently in China particularly in Xinjiang newspapers, and party documents. However, it is always enclosed in carefully quarantined quotation marks and demarcated as a source of danger through its collocation with “The Three Evils” (Dwyer, 2005, p.51-52; Millward, 2007, p.x). Collocation is used here to refer to an inter- textual practice of representation and the production of meaning. Concepts become associated with other concepts in public discourse through the ubiquitous but uncritical repetition of their connections. In Xinjiang, the quotation marks which enclose East Turkestan throughout the party-state’s minzu tuanjie texts 52 and in Pan Zhiping’s (2008) book de-legitimise East Turkestan as an identity from the outset. East Turkestan is “fake” because is collocated with and enclosed within a quote inferring its unproblematised non-existence in a way which does not merit mention. Furthermore, East Turkestan is ‘dangerous’ because it is collocated with “The Three Evils” and dongtu, which appear almost universally alongside one another in these texts so that they become interchangeable. They are understood through one another as equal sources of danger.

These repetitions of collocations represent a religion, culture, and peoples as a “source of threat” (Said, 1997, p.xxi). East Turkestan can be publicly discussed but only as a threat and as something which terrorists would say thus remaining within the unbreakable bounds of quotation marks. “Dongtu” and “The Three Evils” are texts which can only be understood with reference to other texts. How they correspond to reality is a secondary question because that is not their function. As representations, their function is to produce values amongst human beings and impose a specific order on human relations (Shapiro, 1988, p.11). These collocations are part of an identity-security discourse, which police the indivisible internal/external boundaries of identity. This policing criminalises Uyghur-ness internal to the territory of Zhonghua Minzu but discursively situated as outsiders through their association with

52 For example, see: Ministry of Information (2009); XUAR Dept of Information (2009). This also is the case for the more widely accessible White Papers. See: State Council (2002; 2003).

103 the threat of ‘Western’ imperialism and ‘Islamic’ terrorism. This discourse includes and excludes at the same time by demarcating the boundaries of China through the threat of the East Turkestan Other. This demarcation produces relations of power between different subject positions so that Uyghur-ness defined through language and religion becomes associated with danger. Chinese-ness as a national identity is securitised via its relational production against Turkestan as a threat from inside and out.

Language policing is part of the nation-building project which seeks to convert Turkestan into Zhonghua Minzu. The techniques of identity inclusion/exclusion highlighted here can be understood through what Mattern called “representational force” (Mattern, 2005, p.602). The force-wielder threatens victims’ narratives of self by narrating away its ‘truth’, by pointing out inconsistencies in the narrative, and dismissing alternative interpretations of ‘reality’ as false (Mattern, 2005, p.603). However, the party-state is tacitly acknowledging that Turkic-ness is a felt reality for many Uyghurs. The CCP devotes considerable resources to narrating it out of reality by reminding people of the dangerous nature of such a “mistaken” identity. Contemporary Turkic-ness is represented as outside China and outside security by its collocation with “The Three Evils”. Only in the “mistaken understandings” of “The Three Evils”, which threaten minzu tuanjie, can Uyghur history be seen as outside Zhonghua Minzu and can Uyghur identity be framed as linguistic (Ministry of Information, 2009, p.47, 61, & 94):

‘Turkic minzu is only a historical category, today there is no reality to this minzu…lots of different countries use English, can you say these are one minzu? By this logic, if you take ‘Turkic language’ as to call Uyghurs, Kazakhs, Kyrgyz, etc one minzu, this is obviously mistaken’. (Ministry of Information, 2009, p.58)

This is a silencing of a key self-identified feature (language) of how Uyghurs conceptualise Uyghur-ness53. This silencing articulates Chinese-ness by excluding the very idea that Turkic- ness can exist as a form of self-identification. This “representational force” is an attempt to silence Uyghurs self-articulations and a means to identify from above what it means to be Uyghur: a peripheralised member of Zhonghua Minzu ‘outside’ the Han nucleus. This exclusion of self-identified characteristics and re-imagining of Uyghur-ness, a means of conversion, can also be found in how the relationship between Islam and Uyghur-ness is identified for Uyghurs. The enclosure of Uyghurs into China away from the Turkic speaking world and the Islamic world is an important dimension of external boundary demarcation. For example, the Dictionary of History and Culture of Every Xinjiang Minzu explains that users of Turkic languages are distributed across Xinjiang, , and but conspicuously omits mentions of the millions of speakers across Central Asia (A, 1996, p.316). The interpellations of Uyghurs as

53 This identification will be discussed and analysed fully in chapter 6 using interview materials.

104

Chinese and “not a Turkic minzu” are followed in the 50 Whys with a narrative which denies the centrality of Islam to Uyghur identity. Uyghurs are told they are “not an Islamic minzu” because religion is not one of Stalin’s 4 defining principles of nationality54 (Ministry of Information, 2009, p.55). Religions are said to include followers of different minzu and Uyghur history has many religions including of several thousand years ago (Ministry of Information, 2009, p.55-58) 55 . This narrative de-emphasises both the Islamic past and ongoing Uyghur self- identification with Islam. The party-state is narrating Uyghur history through discontinuity where Han-ness is relationally represented through continuity. The multiple origins and religions in this narrative on politically acceptable Uyghur-ness can be contrasted against the performances of homogeneity of Han-ness, associated with nationhood and modernity:

In reality, many Uyghurs are not followers of Islam. For example, members of the Communist Party, members of the Communist Youth League, officials working in Government organs, students, etc are not religious. They are not devout Muslim followers of Islam, yet we cannot deny their Uyghur identity (shenfen56). (Ministry of Information, 2009, p.56)

The party-state is using these examples to redefine Uyghur-ness for their own ends. In the reform era, the party-state has shown some awareness that is impossible to govern complex societies in a monolithic way. “Managed pluralism” offers a model to include alternative organisations in the state but also to place limits on cultural diversity in the name of national traditions (Balzer, 2004, p.237). The China Islamic Association is one such organisation which is co-opted into the state and used to make ‘authentic’ declarations on Islam. The China Islamic Association’s (2006) Patriotic Muslim Coursebook even goes so far as to say that the CCP should be “worshipped” at an equal level to Allah. Of course there are Uyghurs who do not consider themselves Muslims. However, there are no studies which have found any significant numbers who do reject Islam as opposed to practicing it in private in order to stay in employment. The 50 Whys later reminds readers that in the name of “religious freedom” students and employees of the party-state are not permitted to visit the mosque or practice Islam because these “religious activities” interfere with “freedom of choice” and “modern education” (Ministry of Information, 2009, p.103-104). In chapter 2, we saw how the shaoshu minzu category was represented as an obstacle to the formation of modern nationhood in the discourse of “ethnic extinction” (minzu xiaowang). Here the party-state identifies a specific social practice of

54 These were mentioned in Chapter 2: common language, common territory, common economic life, and common psychology. 55 On how the official codification of lyrics in Uyghur classical music () omitted references to religion despite Uyghur musicians explaining their meaning through Islam, see Harris, Rachel (2008). 56 Shenfen is an ambiguous official term for identity as it derives from identity card (shenfen zheng) and is used to mean ‘status’ more than self-identification. It is often avoided in debates on identity in Chinese anthropology (also see footnote 5 below) and does not always imply ‘identification’ (rentong) with the category.

105

Uyghurs as particularly problematic because it interferes with “modern education”. These exclusions are central to nation-building and to relationally articulate that to be Chinese is to be modern and irreligious. This ‘interference’ with modernity has nothing to do with economic development but is about securing the cultural boundaries of the nation.

Pan Zhiping claims that “ are not simply places of worship; they are also fortresses for Muslims to instigate combat” (Pan, 2008, p.149). These assertions inscribe the Uyghur social practice of visiting the Mosque not simply as un-modern but as a threat. Such discourses which represent Islam as culturally inferior and backward need not be supported with evidence or supporting arguments for these stereotypes are so pervasive they are taken as truth (Said, 1997, p. xviii). Identification with Islam and with Turkic-ness in Xinjiang are represented through what David Campbell (1998) called “discourses of danger”. “Discourses of danger” demarcate the boundaries between safe insiders as the referents of security and dangerous outsiders as threats (Campbell, 1998, p.1-3). The exclusionary practices which articulate the mosque and the people who practice Islam as sources of danger relationally articulate a safe inside of Chinese-ness as non-Islamic. At the same time, this discourse includes Uyghurs as party-members and state employees. However, by being prohibited from practicing Islam, this inclusion is selective, provisional, and demands willingness to adopt the Communist atheism, which Uyghurs associate with the Han. This model of nation-building produces a nation which theorises difference as a problem that requires exclusion from or conversion into Zhonghua Minzu for modernity and ‘China’ to unfold. These discourses of danger position threats of internal disunity inside the imperialism of global capitalism. These historical threats are perpetuated and represented through the danger of the contemporary Uyghur “terrorist” paradoxically linked to the outside “terrorist” threat of Islam and the democratic ‘West’. “East Turkestan” is said to be supported by “anti-Chinese” “Western enemy forces” (xifang duidi shili), which include the media and unnamed governments by “using 7-5 to damage China’s image”:

For a long time, Western enemy forces have not ceased to use the plot of “peaceful evolution” against us. Particularly since reform, our economy has transitioned towards the market. This has made the West look upon us with so-called hope. They believe China can only lean towards a market economy to develop and will certainly move towards capitalism…but over the last thirty years of reform, China has not move towards capitalism, on the contrary, Socialism with Chinese characteristics has enjoyed unparalleled success, achieving extreme development and China’s rise and improvement in international prestige are unceasing. (Ministry of Information, 2009, p.18)

East Turkestan is being performed in this discourse of danger as the internal/external boundary which produces an enemy of China in the form of outside Western capitalists (external) in collusion with Turkic nationalists (internal). Here we can see the contingent nature of the securitised boundaries of China which cross the domestic and the international, as well as the

106 territorial and cultural. Contemporary Turkic enemies are a reconfiguration of earlier CCP discourse where the enemies were imperialism from the outside and feudalism on the inside. Reality has been “turned upside down” in the words of Mao Zedong, as the party-state employs the category of Socialism today as a form of identity politics, despite the end of the planned economy, growing inequality, and the privatisation of social services. Official Chinese representations of enemies in domestic and global politics are inseparable and are embedded within one another in these acts of performativity. These discourses define what China is through a “constitutive outside” of capitalism and Turkic-ness. These acts produce Chinese-ness as unique and enclosed from the “West” and Turkic-ness. To paraphrase , if the “West” did not exist, it would be necessary for China to invent it.

East Turkestan is far from an outdated or vague term today. The party-state has codified it and continues to provide it with ongoing relevance through its textual exclusions in response to contemporary Uyghur self-identification. With the establishment of the Xinjiang Uyghur Autonomous Region in 1955, the integration of the Turkic-speaking, sedentary peoples of southern Xinjiang with the nomadic peoples of the north has provided Uyghurs with a more distinct and bounded “territorial frame” for their grievances and political aspirations (Bovingdon, 2004, p.3). This “territorial frame” has then been linked by Uyghurs to the cultural politics of exclusion of Turkic-ness in the PRC and inclusion of Uyghurs as culturally inferior to Han Chinese. China did not create East Turkestan but they continue to mutually constitute each other through official discourse. The party-state’s codification of East Turkestan has enabled resistance or what Judith Butler called the “taking up of the tools where they lie” (Butler, 1999, p.185). Uyghurs (re)perform East Turkestan and Uyghur-ness in alternative ways57. One typical response received from a ‘black’ taxi58 driver to the question ‘where are you from?’ was ‘I am a Turk, Uyghurs are Turks. We should have had our own independent country’. Many students told me ‘We are Turks, we are not Chinese’. One teacher told me ‘ethnic unity is just party nonsense. We are Turks’. Interviewees from all social classes, from getihu to intellectuals, expressed some identification with being Turkic even if it wasn’t their primary form of self- identification. When discussing Turghun Almas’ banned book and its framing of Uyghurs as Central Asian, a young businessman, Muhtar laughed saying ‘this is Central Asia. Han only came here recently’. East Turkestan remains a powerful symbol in the national

57 This research found near universal Othering of the Han and frequent self-identification in contrast to what was deemed Han-ness and Chinese-ness. This is not the only form of identification or boundary which matters to Uyghurs or which scholars of Xinjiang should explore. However, it is one which informants chose to discuss on a daily basis and one which comes to the fore any time politics are discussed. 58 In Urumchi almost all legally registered taxis are owned and operated by Han. Many Uyghur drivers complained of discrimination and their response was to drive a ‘black taxi’ (heiche), an unlicensed car which picks up passengers discretely and is not taxed.

107 imagination of Uyghurs despite official discourse and political restrictions on public articulations of self-identification.

The party-state demarcates the boundaries of Chinese-ness by designating Turkic-ness as outside these boundaries and as a danger to the maintenance of identity-security. However, Uyghurs are resisting this nation-building from within by adopting its language and (re)performing these boundaries in ways which re-assert their cultural difference and position themselves as outside Zhonghua Minzu. Resistance and power are thus being mutually shaped59. Uyghurs take up the tools available to them which takes place within the “orbit of the compulsion to repeat” and agency becomes “located within the possibility of a variation on that repetition” (Butler, 1999, p.185). There are theoretical debates on these topics within Chinese anthropology at least in Beijing, particularly through the Central University of Nationalities60. The position taken here follows Pan Jiao (2008b) that the distinction between ethnicity and nation are more to do with recognition of demands and relations of power than identification or ideas about shared history (Pan, 2008b, p.103 & 112). Yet, as we shall show in the next section, the contingency of the representations of Xinjiang and the policing of the terms Turkestan and terrorism do not simply reflect power. They are central to producing relations of power. These representations denote Uyghur-ness and Turkic-ness as an ethnicity (danger) within Xinjiang and within China, where Chinese-ness is conversely represented as a unified nation (security). However, nation-building and securitisation are unintentionally producing destabilising effects as Uyghurs turn these discourses against themselves.

3.2: “The Three Evils” and the Production of Resistance

This section will explore how the party-state’s discourse of “The Three Evils” of “terrorism, separatism, and extremism” simultaneously perform the theoretically indivisible internal/external and domestic/international boundaries of community. The empirical focus is on how shifts in representations of security and Uyghur-ness have been contingent on political context. As we saw in the introduction, since China declared it’s “war on terror” following the events of September 11th 2001 up until July 2009, Xinjiang had in fact experienced less violence and less incidents of unrest than any other part of China (Bovingdon, 2010, p.112).

59 Chapter 6 will use interview materials to fully discuss how Turk becomes a category of resistance to the power of Chinese nation-building. 60 Of particular note are the debates on whether minzu as a state-recognised category or zuqun as a self- defined form of ethnicity are the most appropriate unit of analysis in social anthropology in China. These debates are very much tempered by the political exigencies of life in China thus they remain relatively abstract, rarely touch upon issues in Xinjiang, and adopt the jargon of the party-state in relating these questions to the promotion of “harmony” and “unity”. For example, see Ruan (2004); Zhang (2006); Ma (2007); Pan (2008b).

108

There is no objective relationship between danger and the event or mode of identity from which it said to derive (Campbell, 1998, p.2-3). Danger does not exist independently of those to whom it may become a threat (Campbell, 1998, p.1). Despite, its relative stability compared to other parts of China, Xinjiang continues to be represented through danger and the “inside/outside Three Evils”. This section will show that these are “discourses of danger” which identify Turkic-ness and Islam as threats to Chinese identity-security. They are central to nation-building because they articulate the boundaries of inclusion by highlighting what identities are to be excluded. By looking at the nature of Uyghur protest in Xinjiang this section will show how resistance and power are not exterior to one another and that they shape each other in their articulations of identity politics.

The concept of the internal/external boundary is used to highlight the indivisibility of inclusion and exclusion. By excluding things as outside Chinese-ness, the party-state articulates what is to be included inside Chinese-ness and vice-versa. A key strategy of the party-state in its demarcation of the internal/external boundaries of the nation emerges in the discourse of “The Three Evils” through the frequent use of the collocating phrase, “inside/outside” (jingneiwai)61. “Inside/outside” is an attributive phrase to link disaffected Uyghurs inside the territory of China, particularly those in section 1 who frame Xinjiang history as distinct from China, with the external dangers of Islam and Central Asia. This frames Uyghurs as physically inside the territory of China. However, they are simultaneously positioned outside the timeless civilisation of Zhonghua Minzu. This is because they identify with Turkic-ness and Islam, which are being framed as un-Chinese. For example, the Regional Government Chairman Nur Bekri’s (2008) speech at the Regional Government Congress reproduced this discourse with constant use of the phrase “the inside/outside Three Evils”62. In a theme continued in the media and all the education sources cited here, he claimed that “anti-Chinese Western forces” (xifang fanhua shili) and Al-Qaeda were supporting East Turkestan terrorists (dongtu). These un-named forces are “promoting infiltrating, harmful, disturbance-instigating, and subversive activities” as part of their overall strategy of “westernisation” and splitting the motherland (Bekri, 2008).

There were numerous incidents of unrest recorded during the 1990s in Xinjiang included riots, protests, assaults on individual security forces, and most notably the Baren 1990 and Ghulja 1997 incidents. The case presented by the Chinese government’s documentation and news reporting on “terrorism” in Xinjiang has been characterised by serious scholars in this area as inconsistent and lacking evidence or independent corroboration63. No Uyghur group publicly supports violent acts (Millward, 2004, p.30). The scholarly consensus among Xinjiang specialists

61 For example, see Bekri (2008); Ministry of Information (2009); State Council (2009b). 62 Since then “inside/outside” has become frequently used in all references to “The Three Evils”, including Ministry of Information (2009), Wang (2009; 2009b), and State Council (2009b). 63 For example, see: Millward (2004); Bovingdon (2004; 2010), and Roberts (2012).

109 is that while these events presented serious challenges to the authority of the CCP in Xinjiang, there is little if any evidence of transnational terrorism: “few if any of these incidents resemble the premeditated, targeted, and substantial acts of violence one usually associates with international terrorist groups” (Roberts, 2012, p.3-4). However, the official representation of these events will tell us much about the contingency of identity-security but also the discursive realities people in Xinjiang need negotiate. These events have been catalogued by academics such as Joanne Smith Finley (1999), Michael Dillon (2002), Gardner Bovingdon (2004; 2010), James Millward (2004), and Sean Roberts (2012). However, the approach here is less to uncover the truth behind these incidents and more to explore how official representations of them produce truth. The party-state’s shifting discourses of Otherness illustrates the contingency of Xinjiang’s place within China. The meaning of internal/external boundaries can be altered but the existence of the boundary remains and an appearance of immutability is maintained. Anthony Cohen argued that people “think themselves into difference” such that categories persist but the meanings of boundaries are contingent and shift with social practice (Cohen, 1985, p.86 & 91). The representations of threats are constantly reconfigured but are retrospectively represented as unchanging.

Slogans shouted and placards waved at demonstrations and incidents of the 1990s in Xinjiang were forms of resistance against the perceived power of the government and the Han Chinese population. They did not call for establishment of an Islamic Caliphate as the above-mentioned Chinese sources assert. For example, slogans such as “independence, freedom, and sovereignty for Xinjiang” and “Hans out of Xinjiang” were the most vocal and typical of these incidents (Bovingdon, 2004, p.7). This included impromptu protests in 1985, which eventually attracted thousands of Uyghurs on to the streets of Ürümchi shouting slogans against in-migration of Han Chinese, birth control policies, and nuclear testing (Millward, 2004, p.8). Ismayil Ähmäd, the then Regional Government Chairman, was subsequently removed from his post and “kicked upstairs” for suggesting Uyghurs should be given employment before more Han Chinese should be allowed to migrate to Xinjiang (Bovingdon, 2004, p.7). One 40 year female member of the Uyghur diaspora in Europe echoed Foucault when she explained the violence of July 2009 by saying “where there is oppression, there is resistance”. Mukhtar, a politicised intellectual interviewed in later chapters, explained the violence of July 2009 in terms of restrictions on political and cultural expression by saying that “if you have a boiling pot of water and you keep the lid on too tight it will eventually boil over”. The “snowballing” nature of violent incidents such as Baren and Ghulja where small scale fights, disputes with officials, and single issue protests can rapidly and spontaneously attract large numbers of protestors suggests more wide- spread discontent in Xinjiang than organised, pre-meditated violence (Bovingdon, 2005, p.7). Most disaffection, as Bovingdon (2002) argues, centres on claims of inequality in the PRC and to cultural distinctions against the Han. This discontent resists the discourses discussed in

110

Chapter 2 where Han are represented as a source of stability and modernity by identifying them as the obstacle to “freedom” for Xinjiang.

Contrary to the constant denials of the early 1990s, it was 1996 before the Party declared it had a problem with “domestic splittists” (Dwyer, 2005, p.52). The Baren “uprising” of 1990 had witnessed a group of Uyghurs in the south west of Xinjiang attack a border police station and steal firearms after a protest in a mosque against family planning had spiralled into a riot (Bovingdon, 2004, p.8; 2010, p.123-125; Dillon, 2002, p. 62-65). The narrative presented by party documents describes the Baren “uprising” as a “turning point” and a “wake-up call” for the CCP (State Council, 2002; Bekri, 2008, section 3). China was then said to progress in linear fashion from being threatened by “separatism” to the events of Ghulja/Yining in 1997, which are represented as the victory against the “splittists” (State Council, 2002). The Baren uprising sparked a reversal in the post-cultural revolution liberalisation of speech in Xinjiang as mosque closures and arrests of Imams increased (Bovingdon, 2010, p.87-88). For example, security services conducted mass seizures of Uyghur music tapes and all lyrics then had to be vetted by a CCP committee before being recorded or performed (Bovingdon, 2010, p.94).

The Ghulja/Yining incident of 1997 is described by US scholars and Uyghur diaspora groups as a massacre of Uyghurs protesting for the right to promote education on alcoholism through the medium of traditional Uyghur gatherings (mesrep). These sources, which rely on Uyghur testimonies claim that Uyghur protestors were rounded up indiscriminately by security services, forced into the local football stadium, and hosed down in sub zero temperatures64. Police fired on protestors two hours into a peaceful march for the release of one of the arbitrarily detained leaders of the Mesrep ‘movement’ (Bovingdon, 2010, p.125-126). These events were a turning point because by 1998, General Secretary Jiang Zemin announced that a “Strike Hard” campaign was necessary as part of the “long term task to fight splittism” (quoted in: Dwyer, 2005, p.52). The contingent shifts in representation are concealed in this discourse where separatism and struggle became imagined as integral components of Chinese nation-building throughout the history of Xinjiang. Uyghurs became represented as separatists, which sits uneasily alongside the folklorised images of happy dancing shaoshu minzu. These shifting representations reflect the contingency of security and danger in Xinjiang and the identities which they articulate. Where there was no insecurity there was now a “long-term” struggle against threats to the existence of the hitherto united Zhonghua Minzu. This historical re- packaging has become seamlessly incorporated into the party-state’s explanation of the minzu problem as a “long-term struggle” against “The Three Evils” as a remnant of earlier stages of

64 For example, see sources which use local Uyghur testimonies on the incident: Dillon (2002); Starr (2004); Millward (2007).

111 historic-cultural development65. These shifting security discourses perform different Chinas by performing different Xinjiangs.

Despite the contingency of the identities these security discourses perform, the Ghulja/Yining incident has today been re-represented as the “victory” of a “long-term struggle against terrorists” (Bekri, 2008, section 3). The State Council’s (2002) document, East Turkestan Forces Cannot Get Away with Impunity, claimed that only “imperialists” and “separatists” could possibly use the term East Turkestan in accordance with the discussion of in section 1. Jiang Zemin’s 1996 statement made no mention of religion but after the events of 9/11 “splittists” became repackaged as “religious fundamentalists” (quoted in Dwyer, 2005, p.52). Rather than blaming “domestic splittists”, the State Council retrospectively explained all the incidents of the 1990s as part of “international terrorist forces” “inside and outside Chinese territory” (State Council, 2002). The Chinese Foreign Ministry announced in 2002 that “we should be cracking down on these terrorists as part of the international struggle against terrorism” (Dwyer, 2005, p.52). The “Strike Hard” campaign’s stated objective changed after September 11th to the defeat of “The Three Evils” of “terrorism, separatism, and extremism” and part of the global “war on terror” (Starr, 2004, p.15; Bekri, 2007). These contingent shifts in signification have enabled a re-representation of disaffected Uyghurs as “Islamic terrorists” (Starr, 2004, p.15; Smith, 2006, p.132). The “smiling, patriotic Uyghurs and Tibetans of the 1980s” have given way to the disgruntled, “backward” terrorists of the late 1990s” (Gladney, 2004, p.364). The internal/external boundaries of Chinese-ness once again shifted with representations of Xinjiang from peaceful to “domestic splittism” and now to “international terrorism”. With the re- representations of ‘outside’ threats, the ‘inside’ of China is being re-represented as insecure in shifting ways. The party-state’s shifting discourses of Otherness illustrates the contingency of representations of Xinjiang and its place within China and the world.

Here we see how states can superimpose new categories from above onto older categories reconfiguring rather than transforming difference. Prior to 9/11 and the invasion of Afghanistan when “Islamic terrorism” replaced the Soviet Union as the gravest threat to the security and modernity of the ‘West’, the party-state’s texts produced a reality of “domestic splittism”. By January 2002, there was suddenly a timeless threat linked to the danger of the Islamic world such that authors such as Pan Zhiping could write without supporting evidence that “In China’s Xinjiang and outside its borders sees the forces of the scum Bin Laden with the evil ‘East Turkestan’ terrorist forces threaten the security of the life and property of the nation and the people” (Pan, 2008, p.2). “Terrorism” is a powerful discursive category which automatically de- legitimises resistance and political opposition particularly in the global context of the “war on

65 On the “long term” nature of the minzu problem, see: Ministry of Information (2009), XUAR Dept of Information (2009), and Pan (2008).

112 terror”. However, the appearance of continuity remains as history becomes re-written. The timelessness of the nation is understood through internal/external boundaries which are threatened by the timelessness of “The Three Evils” bringing dangers from Central Asia and the Islamic world into China.

The shift in representation of Uyghur discontent from “domestic splittism” to “international terrorism” can be understood as an appeal to the growing anti-Islam discourse following the collapse of the Soviet Union, which Edward Said (1997) reviewed in Covering Islam. However, its timing immediately after 9/11 and its methods explored below suggest this was an exploitation of geopolitics for domestic ends. The State Council’s (2002) document was circulated across Chinese mainstream media including the main pages of Xinhua news and translated into English for international media. The party-state effectively seized on the opportunity to gain international legitimacy for its nation-building project in Xinjiang. This served to obscure ethnic tensions in the region as the US and the UN recognised the hitherto unknown East Turkestan Islamic Movement (ETIM) as a transnational terrorist organisation linked to Al-Qaeda (Roberts, 2012, p.3). The re-representation of boundaries claimed that Xinjiang had long been struggling against organised terrorist groups linked to Al-Qaeda and other transnational terrorist networks such as the Islamic Movement of Uzbekistan and even Hizb ut-Tahrir (State Council, 2002; Bekri, 2008). China was seeking international legitimacy for its ‘domestic’ nation-building project in Xinjiang by locating its efforts in the global “war on terror”. This can be called “harmonising the world” into the new “benevolent rule” of Chinese order (Callahan, 2012). However, this harmonisation is focused not on attaining international power but is used to gain support for the party-state’s strategies to resolve domestic problems in its confrontation with its own frontiers.

The State Council’s (2002) document claimed that “There is plenty of evidence to show that most of the terrorist and other violent incidents which have occurred in Xinjiang were directly plotted and engineered by the “East Turkistan organisation” (dongtu) beyond China's borders, with the collusion of a handful of people within the borders”. The complete absence of evidence cited is typical of these policy statements. However, this does not hinder the production of an internally consistent narrative on nation-building, which seeks to selectively include Uyghurs assuming they do not identify with Turkic-ness or Islam. Since 9/11, ‘East Turkestan’ has become understood through this collocation. Dongtu refers interchangeably and simultaneously both to ETIM, the purported terrorist organisation, and ‘East Turkestan’ as an idea 66. This collocation is a form of boundary production and identity policing because it becomes impossible to discuss ‘East Turkestan’ (dongtu) in Mandarin Chinese without it containing the

66 ‘Dongtu’ is employed in this manner in all the official texts used in this chapter so that an uncritical reading renders ETIM and East Turkestan as synonymous.

113 reference to the terrorist organisation ETIM (dongtu). It is technically possible to use the full Chinese name for East Turkestan (dong tujuesitan). However, given the practice of shortening long transliterated names in Mandarin Chinese, all interviews with Han residents referred to dongtu not the full name. Even when I used the full name, interviewees responded using the shortened version interchangeably to mean ETIM and East Turkestan. These discourses and their collocations are forms of securitisation. The challenge which Uyghur discontent presents to Zhonghua Minzu was once blamed on imperialists from the outside but is now blamed on the Islamic world, which is both outside and inside China at the same time. Zhonghua Minzu is officially positioned as a realm “beyond politics” where the Islamic and Turkic outside are existential threats to China’s sense of “we-ness” (Buzan et al, 1998, p.4; Wæver, 1995, p.60). Securitisation does not emerge from an objective sense of we-ness. Rather it is a performative enactment of the “we-ness” of Zhonghua Minzu. These enactments of identity-security perform the contingent and shifting boundaries of nation it claims to simply reflect. In this case, Zhonghua Minzu is the “we” to be secured and the threat of the Islamic world outside China in collaboration with the Turkic world within is that which is to be secured against.

As Winichakul (1994) showed, linking people or peoples within the state’s borders who challenge the official conceptualisation of nationhood to external threats is a key strategy of security discourses. This transformation of the internal and disagreeable into the external simplistically represents national identity as singular, homogenous, and unified in order to promote agendas of social control and nation-building (Winichakul, 1994, p.169-170). In the case of official representations of Xinjiang the danger is articulated as the Uyghur (inside) in collaboration with “Islamic terrorists” and the older enemy of “Western” imperialism (outside). Just as dongtu becomes impossible to discuss without implications of threats from ‘outside’ China, the Uyghur is paradoxically highlighted as foreign. By targeting particular forms of Uyghur-ness as security threats because they are linked to forces outside China, the nation- building project continuously highlights their linkages to cultural forms deemed un-Chinese. The Uyghur is officially performed as inside and outside Zhonghua Minzu at the same time. Their inclusion is based on the extinction of the cultural characteristics they identify with (language and religion).

The representation of Xinjiang as a source of danger to China and to the world, which relationally produces China as source of security and world peace was exemplified in the case of the Uyghurs imprisoned in Guantanamo Bay. The Guantanamo Uyghurs were identified by the CCP as members of the East Turkestan Islamic Movement (ETIM) inside China but linked to Al- Qaeda outside (Roberts, 2012, p.8-9). Interrogation and subsequent interviews by US security services of the Guantanamo Uyghurs discovered they had never heard of Osama Bin-Laden until after they arrived at Guantanamo (Roberts, 2012, p.9). They had only ever fired one gun

114 in their lives, and while they declared the Chinese as “their enemies” they expressed support for the United States and not for Pan-Islamism (Roberts, 2012, p.9-10). That these Uyghurs were released by the United States because they had no links to Al-Qaeda and subsequently re- settled in Albania is not reported in China (BBC, 2006; NYT, 2007). However, for a Chinese audience, Dongtu as discourse serves to securitise the internal/external identity boundaries of China. Dongtu conflates East Turkestan as a cultural concept and as international terrorism where Uyghurs are a threat both inside and outside China.

By linking ETIM (dongtu) to Al-Qaeda Uyghurs are not simply represented as a threat to Chinese rule. They are also represented as a threat to global security because they are enmeshed in transnational networks of “Islamic terrorism”, which seek to establish an Islamic caliphate across the world. China is thus positioned as a representative of global security in its struggle against “the inside/outside Three Evils”. During August 2011, the Foreign Ministry and Kashgar city officials blamed an outbreak of protests and violence in Khotan and Kashgar in August 2011 on Pakistan-trained members of the East Turkestan Islamic Movement (ETIM) (Zenn, 2011). The following month, a spokeswoman for the Autonomous Regional government admitted attackers had been loosely organised with no evident links to “extremists” in Pakistan or elsewhere, and used only “homemade weapons” (Spegele, 2011). The Foreign Ministry refused to comment on the details of the case. However, it continued to link local grievances inside China to international security and the external threat of “terrorism” by saying that, "in Xinjiang and Central Asia, there have been a series of violent terrorist incidents in recent years. They have all involved domestic and foreign collaboration” (Zenn, 2011). These are incidents which characterise the unsubstantiated assertions made by official Chinese sources but which produce a social reality of security and danger which are inseparable from identity politics67. The framings of these incidents blur the internal and external boundaries in the production of China and its enemies by highlighting the need for nation-building because of the foreign-ness and un-Chinese-ness of Uyghurs.

The party-state’s re-representation of these incidents as part of the identity-security threat of “The Three Evils” represents Uyghur discontent not simply as an issue of public order but as a security threat to existence of China itself. Uyghur discontent is being criminalised much as all overt resistance to party-state policies across China but it is being imbued with meanings which mark Xinjiang off from China as a site of Otherness and danger. The “Strike Hard, Maximum Pressure” campaign was initiated in 1996 to counter growing social unrest in the region during the turbulent 1990s. As part of Strike Hard, the party-state has criminalised those literary works which locate themselves out-with the Chinese sphere or present a gloomy representation of

67 The final chapter will fully explore how these ‘realities’ are negotiated and (re)produced by Han Chinese and Uyghur residents of Ürümchi.

115 contemporary life in Xinjiang as threats to national unity (Friederich, 2007, p.95). The strategic conflation of ordinary criminals, political activists, and writers as “separatists” and later as “terrorists” was exemplified in the Party’s 2002 slogan “wipe out pornography and strike at political publications” (saohuang jizheng) alongside the public burning of pornographic materials and books on Uyghur history (Becquelin, 2004, p.45). The criminalisation of speech in Xinjiang is part of the broader de facto policy shift from censoring writers in the 1990s to imprisoning them in the 21st century under the auspices of what Jiang Zemin referred to as “terrorism in the spiritual form” 68 (Bovingdon, 2010, p.99-100). Nicholas Becquelin (2004) described the conflation of peaceful and violent groups under the appellation of “East Turkestan Terrorist Forces”. This conflation is a means to use the “spectre of terrorism” to silence opposition to monolingual language policies and restrictions on religious practices, which “criminalises” Uyghur ethnicity (Becquelin, 2004, p.45). This closes the public space available for Uyghurs to express dissatisfaction lest they be labelled a threat to national security.

One author who transgressed the rules of how public space is constructed by the party-state was the (now) famous Uyghur historian Turghun Almas. Turghun Almas remained under house arrest from 1991 for publishing the book Uyghurlar until his unexplained death on September 11th 2001. Uyghurlar claimed Xinjiang has always been part of Central Asia, thus locating Uyghur history in a framework of Turkic history. In a claim to indigeneity, he argued that Uyghurs lived in Xinjiang prior to the establishment of Han Dynasty colonies (Rudelson, 1997; p.155-157; Millward, 2007, p.344). Resistance to Chinese rule is best understood within the epistemic violence of identity discourses and the physical violence of politics in Xinjiang. Official Chinese claims which perform the security of Xinjiang as an inalienable and timeless component of Zhonghua Minzu are being inverted and resisted through claims to cultural and political connections to Central Asia and Turkic-ness. A more recent case study which illustrates both the criminalisation of Uyghur self-identifications and the inter-related nature of power and resistance is that of freelance writer Nurmuhemmet Yasin. Nurmuhemmet Yasin was arrested on the 29th November 2004 and convicted to 10 years imprisonment in Xinjiang No.1 Jail, a re- education through labour prison camp (laogai), at a closed trial on February 2nd 2005 (CECC, 2005; Laogai Research Foundation, 2008; PEN, 2009). Nurmuhemmet’s arrest followed the publication of his allegorical short story Wild Pigeon (Yawa Kepter) in the autumn 2004 Kashgar Literature Journal. All copies of the journal were then recalled by the authorities (CECC, 2005). The editor of the journal, Kuras Sayin, was subsequently jailed for three years for allowing it to be published (Bovingdon, 2010, p.99). The story tells the tale of a wild pigeon who chooses to commit suicide rather than sacrifice his freedom to live under the yolk of humankind and their hordes of “tamed pigeons” (Yasin, 2005; Laogai Research Foundation, 2008; PEN, 2009).

68 On the arrests of the poet Tursunjan Amat and historian Tohti Tunyaz for researching Uyghur history, see Bovingdon (2010) chapter 4.

116

Nurmuhemmet is described in the campaign for his release as a “writer of Children’s Stories” but by being charged with “inciting splittism” (AI, 2012; CECC, 2005), he and his work is criminalised as part of the identity-security of the “Three Evils”.

The story Wild Pigeon is a good example of how nation-building is producing resistance in Xinjiang. The Uyghur historian, Dolkun Kamberi, translated Wild Pigeon into English. He explained that it had met “widespread acclaim among the Uyghur people” because of its “strong portrayal of a people deeply unhappy with life under Beijing’s rule” (Yasin, 2005). The story is allegorical and echoes widespread concerns amongst Uyghurs about the loss of community, Uyghur language, and freedom under party-state rule. A youthful wild pigeon is told by his elders that “the soul of the entire pigeon community has already disappeared” but they cannot explain why. This articulates Uyghur concerns of what they see as assimilation under Chinese nation-building. One interpretation of the story was that the wild pigeons represented those Uyghurs from across the border in the Central Asian states outside the Chinese sphere (Bovingdon, 2010, p.100). However, the story can also be seen as a commentary on the Uyghur-Chinese boundary within Xinjiang. The wild pigeon is unable to understand those pigeons which have been tamed by humans because they speak their own “mother-tongue”. This confusion reflects and performs the Uyghur concern of loss of mother-tongue under the party-state’s move to Mandarin-medium education69. The humans keep pigeons in a “cage to feed us”. The tamed pigeons say this is right yet “no pigeon among us is permitted to object to this arrangement”. The tamed elders tell the wild pigeon that they should be “satisfied with what they have”. However, the wild pigeon describes the cage as “supremely clever in its cruelty” because it allows “ample view of the freedom denied to him – with no hope of regaining them”.

The inequality between pigeon and humans is a commentary on the inequality between Uyghurs and Han in contemporary Xinjiang. When the wild pigeon says “by caging my body, they hope to enslave my soul”, inequality is not framed as something material but intimately related to identity. The cage is an attempt to eliminate Uyghur culture through the forcible and hierarchical inclusion of Chinese nation-building where pigeons can eat as animals but can only gaze upon the freedom enjoyed outside by humans. The wild pigeon’s mother represents what Jay Dautcher (2009) referred to as the chthonic identity of Uyghurs by telling the young pigeon “they want to chase us from the land we have occupied for thousands of years…they want to change the character of our heritage…strip us of our memory and identity”. The pigeon’s mother represents resistance to the discourse of nation-building that Han have occupied Xinjiang since ancient times. She resists the humans as outsiders who seek to convert Uyghurs by re-writing history and performing Uyghur identity as under threat from this process.

69 This topic is the subject of extensive interviews in chapter 6 and 8.

117

“For the sake of our motherland”, the young pigeon is released by his mother but is later captured by humans. The articulation of Uyghur-ness and identity-security is most obvious when the wild pigeon is then given the choice of freedom or death by his mother who gives him a poisoned strawberry. The pigeon chooses to “die freely” and “restore the honour of our flock”. The identity of Uyghurs is thus performed as in captivity and facing the existential threat of assimilation. Identity like security is thus a matter of life or death. The story is a form of resistance to Chinese nation-building because it represents Uyghur identity as under threat from the writing of history and the language policies of the party-state. It resists power but uses these discourses against themselves by responding to the timelessness of Zhonghua Minzu with performances of chthonic Uyghur identity, which claim Uyghurs occupied Xinjiang before the Chinese arrived. The direct political effect of these forms of “everyday resistance” may be impossible to quantify but it keeps the idea that Uyghurs are a nation outside Zhonghua Minzu in circulation (Bovingdon, 2010, p.205). If we understand identity more of an effect than a cause of social practice, everyday resistance is one of the practices through which identity comes to be taken as a real thing by those who understand themselves through it.

Intellectuals enjoy perhaps more freedom in China today than at any time in its modern history. Novels such as Chan Kuanchong’s The Fat Years (2011) exemplify how resistance to the politics of the party-state is more tolerated in public spaces of Han-populated cities where identity politics are less subject to security discourses than Xinjiang. In The Fat Years, Chinese citizens are drugged by the party-state to make them happy and content. In the opening of the story, one character raises the matter to be told by his friend “just forget it…just look after yourself” (Chan, 2011, p.27). This suggestion to pursue only self-interest and avoid politics articulates the concerns of loss of community and social responsibility across China in the reform era. called this Beijing’s “pretend smile” where people’s external happiness hides a corrupt political structure and a lack of social responsibility (CNN, 2007). The Fat Years was banned in but it was not subject to criminalisation and securitisation. The story of the wild pigeon being told by tame pigeons not to disrupt the existing social order echoes some of the key themes of The Fat Years. The securitisation of Chinese identity and the criminalisation of Uyghur-ness in Xinjiang are such that even allegorical writings are criminalised if they express alternative self-identifications to the official understanding of Zhonghua Minzu.

Overtly and covertly political resistance in Xinjiang is being framed as a threat to China itself thus it is being met with prison sentences for splitting the nation. Dissent in contemporary China is tolerated when viewed as local or private but when movements such as Falun Gong go national they are equated with sabotage of China (Shue, 2004, p.41). However, what makes it near-impossible to publicly address social problems in Xinjiang is that all dissent is framed

118 through this prism of national security and Uyghur identities become a threat to be eliminated. Furthermore, the bracketing of Uyghur dissatisfaction with the party-state and alternative understandings of history as external functions as a coercive mode of internal identity management, which crosses domestic and international politics. In the case of the Chinese party-state in Xinjiang, the politically acceptable boundaries of the Chinese community are being drawn by linking internal discontent with “the outside”. Security and danger, like Self/Other and inside/outside, can only be defined relationally and these dichotomies are mutually reinforcing. These securitisation performances link Uyghur-ness to “international terrorism” thus performing Uyghur-ness and Chinese-ness into global politics as a security threat and a security enforcer respectively. Uyghurs are internal and external at the same time for while they are subject to exclusion, they are simultaneously integrated into China as an indivisible component of a timeless, multi-ethnic nation. These discourses articulate internal boundaries because they perform cultural characteristics as befitting and unbefitting of the nation. They are also external boundaries because they demarcate a world of danger outside the Chinese nation infiltrating the inside.

3.3: The Securitisation of Modernisation: Boundaries within Boundaries within Boundaries

The final section of this chapter will explore how the internal/external boundaries are used by the party-state to draw spatio-cultural boundaries within Xinjiang between more and less secure and more and less modern regions. This spatio-cultural boundary drawing is ethnicised by highlighting Uygur populated areas as particularly dangerous. The demarcation of these boundaries is linked to the “constitutive outside” through discourses of the “inside/outside Three Evils” which articulate Islam and “the West” as security threats. The indivisibility of the internal and external boundary discussed in earlier sections is further exemplified when we look to how this spatio-cultural boundary drawing within Xinjiang as part of Chinese domestic politics becomes linked to discourses of “terrorism” and international Otherness. Most of the incidents listed by the party-state are centred in Kashgar and other Uyghur-majority areas in the south of Xinjiang such as Khotan, Aksu and Kucha. Xinjiang, south of the Tianshan mountains, largely consists of the extremely arid Tarim basin and is colloquially referred to by Han Chinese in Xinjiang and in party documents as the ‘southern frontier’ (Nanjiang)70. Uyghurs are generally understood to inhabit Nanjiang whereas the grasslands north of the Tianshan, which roll into both Mongolia to the East and present-day Kazakhstan to the West, are understood as the home of Kazakh and Mongolian “nomadic” ethnic minorities.

70 All interviewees, academics, and Chinese language teachers I encountered used this spatial division and these terms both informally and in formal situations such as public lectures, language classes, and banquets.

119

In official documents, the spatial division of Xinjiang into north and south is generally used to designate the latter as a less developed Uyghur region with problems of separatism71. This spatial division was exemplified in former regional party chair ’s comments in a press conference at the end of April 2009. Wang announced that “we must fully realise the strategic value and actual meaning of accelerating the development of Nanjiang”, “Khotan and Kashgar are on the frontline in the struggle against separatism. Guaranteeing stability and the mission of defending and consolidating the frontier are heavy tasks” (Wang, 2009). The rest of the speech is devoted to “modernisation” policies such as the expansion of Mandarin-medium education and demolishing Kashgar Old Town to promote “scientific development” and overcome “backwardness”. This illustrates how “modernisation” and securitisation are mutually constitutive of one another. They are inseparable strands of an internal Orientalist discourse which constructs Xinjiang and Uyghurs through Otherness as backward threats in need of conversion by the party-state’s nation-building project. As the majority population in the South, Uyghurs are being represented as “backward” and in need of conversion through “modernisation” or conquest through the violence of “criminalisation” if they resist. This integrates Xinjiang as a “frontier” in the Chinese nation but not yet of it. The Shule County government released a document in May 2009 which echoed this model of nation-building. It stated that the party-state must continue to use media, film, television, and education to promote minzu tuanjie against “The Three Evils”. Furthermore, “in order to intensify the ideological leadership of the struggle against separatism, Shule County is taking the scientific development concept as its leader” (Shule County Government, 2009). This vision that development will scientifically resolve the ethnic ‘problem’ is recurrent in the party’s discourses on Xinjiang. In 2008 Regional Government Chairman, Nur Bekri, explained that:

In Xinjiang – a remote, relatively backward ethnic minority area with complicated regional and border circumstances – only by ceaselessly quickening the pace of economic and social development and greatly improving the people's life thereby allowing each ethnic group to enjoy the rewards of economic reform.…can the centripetal force of all ethnic groups be strengthened towards the party and government from the foundation, and identification with the socialist motherland. In addition, only in the above circumstances can we reduce the breathing space for “The Three Evils”, consolidate the foundation in the masses of the party, and realise lasting peace and stability. (Bekri, 2008)

Nur Bekri explicitly draws a causal relationship between the economic and the social and between development and self-identification. As Uyghurs level of development advances they will supposedly come to identify with both the Socialism of the party and with the Chinese cultural sphere (zuguo). Bekri’s speech is embedded in the party-state’s ethnocentric theory of cultural evolution, whereby development and identity are intertwined facets of evolution moving

71 For example, see: Bekri (2008); Ministry of Information (2009).

120 from the backwardness to modernity and from ethnicity to Chinese-ness. “Separatism”, ethnic identity, and conceptualisations of the self out-with the Chinese historical and cultural sphere all become markers of economic and social “backwardness”.

Yang Faren of the Xinjiang Regional Party Political Research Office in his book The Western Development Project and the Ethnic Question similarly explained how separatism only appears in less developed, rural, ethnic minority regions72. The local government in Shule County in the South West of Xinjiang released a document in April 2009 stating that as part of the fight against “The Three Evils” of “separatism”, “extremism”, and “terrorism”, all levels of government must now promote a patriotic, collectivist, socialist, and minzu tuanjie education system using Mandarin Chinese as the medium of instruction (Shule County Government, 2009)73. These connotations are implicitly directed towards Uyghurs because the “backward” areas where “separatism” is found are listed as Kashgar and Khotan in the south-west of Xinjiang, which are Uyghur majority cities 74 . The north is thus represented as relatively prosperous and secure. However, the south continues to be represented as an arduous frontier where Uyghurs remain in need of conversion and evolution. The disparity in the living standards enjoyed in these regions is easy to demonstrate. However, these “facts” by themselves show nothing and the link made “backwardness” and security is a form of representation which produces “facts”. These links are central to the performance of ethnic and spatial ordering in which Uyghur-ness is the danger while Chinese-ness represents security. Here we see the emergence of internal boundaries within internal boundaries (Uyghur-ness within Xinjiang within China) embedded in discourses of external dangers. The party-state demarcates and produces the safe, modernising, Han-populated, urban inside-ness of Ürümchi against the dangerous, backward, Uyghur-populated, rural outside-ness of Southern Xinjiang, particularly Kashgar, the centre of Uyghur Islam.

In the party-state’s nation-building narrative, people are either on the side of historical progress and modernisation or left behind. The Three Evils’ (san gu shili) opposition to the “sacred mission” of national unification is said to run “against the flow of history” and “modernisation” thus it will “inevitably lose” (Ethnic Unity Education Board 2009, p.60). The party-state’s discourse on the security threat of ‘The Three Evils’ reveals less concern about armed organisations and more about people’s self-identification and interpretation of history:

72 There is little if any evidence to suggest this is the case and the riots in Ürümchi which were officially attributed to the Three Evils would suggest otherwise. 73 This text, which in many ways embodies the mutually reinforcing dichotomies of Hanzu/Shaoshu Minzu, modern/backward, and security/danger is now used within classes on scientific development within Xinjiang Universities. 74 For example, see: State Council (2009b); Ministry of Information (2009); Bekri (2008; 2009b).

121

The Three Evils believe that under the rule of the Qing, Xinjiang became a Chinese colony. This ignores facts and distorts history. After the establishment of the Western Regions during the Han dynasty, every ethnic group identified with the rule of the central government. (Ministry of Information, 2009b, p.47-50)

The boundaries of what it means to be Chinese and what it means to be Chinese-Uyghur are demarcated such that everything outside the party-state’s narrative is a form of “terrorism, separatism, and extremism”. “Mistaken” ethnic identities, defined by reference to minority languages, it is claimed only exist due to this dangerous influence of the “inside-outside” threat of The Three Evils (Ministry of Information 2009b, p.55-59). The party-state is constructing inside-ness by reference to the “constitutive outside” but this cultural outside is inside the territorial boundaries of China. Uyghurs who interpret the arrival of Qing armies as invasion rather than liberation, identify themselves as Turks, or argue for the promotion of Uyghur language education are excluded from the national community by being linked to a threatening past of backwardness and a dangerous outside of security threats. These mutually reinforcing dichotomies define the boundaries of community. They attempt to symbolically construct national community by imbuing historical representation, self-identification, and language use with normative meanings positioned in a social hierarchy.

Figure 3.1: “Bilingual Education” Hastens Buds (Shuangyu Cuibeilei). This photograph also appeared in the and on the website of the national network with the caption “I can fly” (Wo neng fei).

122

The photograph in figure 3.1 appeared in a free of charge, party-state organised exhibition in the regional expo centre in Ürümchi, September 2009. This exhibition showcased examples of “ethnic unity” (minzu tuanjie) following the riots of July 2009 and the “modernisation” (xiandaihua) of Xinjiang under the rule of the party-state. The exhibit above was displayed in the section on “improving the people’s livelihood” alongside photographic representations and statistics on the development of cotton field production and the oil and gas industries. Its location in the exhibition illustrates how ‘Bilingual Education’ (shuangyu jiaoyu)75, or education using Mandarin as the medium, is being promoted within the party-state’s discourse on development and “modernisation” and is being disseminated to a broad audience. The caption “Bilingual Education hastens buds” suggests that by being educated using the medium of Mandarin, young Uyghurs will be driven to blossom and prosper through the guidance of the party-state and its bilingual education policy. By omitting mention of the young students’ mother-tongue in the exhibition, the implicit suggestion is that this does not offer the same opportunities for personal growth and prosperity. Furthermore, the caption “I can fly” articulates Mandarin as a modern language of development and mobility and Uyghurs in need of being developed by it. In Xinjiang, Mandarin Chinese is central to the party-state’s nation- building model. It is being mobilised by the party-state as a symbol of the modern, national community whereas minority languages are being placed outside of the national community to remain in the ethnic and domestic spheres.

Ma Rong, one of China’s most politically influential social scientists, argues that ‘bilingual education’ is essential for China to “have the ability to develop into a modernised nation” (Ma, 2009, p.240-241). Discourses on Uyghurs are being mobilised to project China’s ‘backwardness’ at the international level as evidence of the need for a strong and ‘modern’ nation. In 2004 the Xinjiang Ministry of Education announced a gradual shift to what it termed Bilingual Education (shuangyu jiaoyu). The plan declared that by 2012 every school in the region was to adopt this system of using Chinese as the medium of instruction in all courses, with the exception of approximately four hours a week of minority literature studies for ethnic groups other than the Han. While this was a new policy, it is better to characterise this as a quantitative rather than a qualitative shift. The Regional government had long echoed the party’s push to promote Mandarin across the country at the expense of other dialects and minority languages. In 1992, experimental bilingual classes were established in cities across Xinjiang to produce minority students capable of fluency in both Mandarin and their native language (Han-min jiantong). The Xinjiang Regional Government issued an “eradicate illiteracy circular” (saochu wenmang tiaolie) in 1996 (XUAR, 1996). Article one stated that illiteracy must be eradicated for the “needs of Socialist modernisation”. Article 10 defined literacy for urban-dwellers as the ability to use 2,000 Chinese characters while article 4 provided the right of the regional government to inspect and

75 On ‘bilingual education’, see Schluessel (2007; 2009).

123 fine work units (danwei) whose members failed to reach these standards. Nevertheless, the 76 million RMB in funds made available for the new drive for Bilingual Education by the central and regional governments show that this now a serious priority for the party-state (Ma, 2009, p.213).

The State Council’s white paper, Development and Progress, tells readers that “practice proves” that “bilingual education” promotes “development, ethnic unity, and harmonious ethnic relations” such that education of minorities is a “national mission” for development and unity (State Council, 2009b, p.19-20). Shen Jianhua, the Head of the political research institute for the CCP in Xinjiang wrote that the promotion of education in Chinese (Hanyu jiaoxue) raises the overall “quality” (suzhi) of ethnic minorities, which is essential to “ethnic unity” and identification with the Chinese nation (Shen, 2009, p.14). The party-state’s conceptualisation of “Modernisation” and its relationship to language-use is being tied to its identity-security discourses, which aims to produce the boundaries between the secure inside and the dangerous outside of China and Xinjiang. Nur Bekri, the chair of the autonomous regional government, has stated in a number of interviews, including for the China Daily in April 2009, that teaching Mandarin Chinese and the policy of “bilingual education” are at the “front-line in defeating terrorism” and developing Xinjiang (Lam, 2009). The 50 Whys tells readers that only “separatists” and “terrorists” oppose the party’s policy of “bilingual education” to institute Mandarin as the sole language of instruction (Ministry for Information, 2009, p.90-93 & 107). The party-state’s ethnicised approach to Modernisation and nation-building are being securitised such that Uyghur language use is being produced as a security threat. Here we see the “constitutive outside” is a performance of an identity-security threat which excludes particular groups within the territorial confines of the nation as ‘outside’ the cultural boundaries of the nation. This identifies Uyghur-majority areas within Xinjiang within China as threats. “Modernisation” policies are part of the projects to promote the “political stability” of existing national borders but the transformation of cultural boundaries within these borders. Nation- building seeks to convince Uyghurs that by identifying with Chinese culture and modernisation, defined through language and demarcated through national borders, their security and prosperity will be guaranteed.

It is impossible to explain the promotion of Mandarin Chinese as a pragmatic or cost-effective means of inter-ethnic communication. In Xinjiang Uyghur and Kazakh languages are mutually intelligible. Uyghur language already serves as an inter-ethnic lingua franca among Turkic speakers and many Mongolians in the region (Dwyer, 2005, p.13). Nevertheless, these groups are expected to learn Mandarin Chinese, a language from a different language family altogether, to be able to communicate with one another. The political drive to produce identification with the Chinese nation and its development model is being prioritised over any immediate

124

‘economic’ concerns but remains embedded in this modernisation discourse. In the first article of The PRC and Script Law (2001) the standardisation and universalisation of Chinese language and script are stated as requirements for inter-ethnic communication and “healthy” social and economic development. The concept of “health” here is projected onto the existing hierarchical ordering of languages to frame non-Mandarin Chinese speakers as a threat to the health and security of the nation 76. This securitises a particular conceptualisation of development and the nation to exclude languages other than Mandarin as outside of Chinese- ness and a threat to these cultural boundaries.

Article 5 of these documents claims that the promotion of Mandarin Chinese is “beneficial for the protection of national sovereignty and dignity, the unification of the motherland, ethnic unity, and the building of material socialist civilisation and spiritual civilisation”. This ‘modernisation’ offers neither a pragmatic economic policy nor a practical means of communication. Language use is being tied to the mutually reinforcing dichotomies of modern/backward and security/danger. Languages other than Mandarin are being framed as ‘outside’ the nation and something unhealthy to be converted or eliminated. National sovereignty here is not something which reflects existing identity boundaries. It is being performed through these discourses of danger as a means to produce and naturalise the relationship between territory and people’s self-identification (Weber, 1998, p.20). These boundaries physically and culturally demarcate the inside of China from the outside of states external to these modern borders and the everyday practice of Turkic language use within these borders.

A shared language is fundamental to any nation-building project. Consuming the literature and identifying with textual representations which produce what Benedict Anderson (1991) termed an “imagined community” depend on a common language. Understanding these representations is not a sufficient condition for the formation of a community of people who have never met but it is a necessary one. However, the promotion of a shared language in Xinjiang is not promoted to produce community per se. It is about the reproduction of a particular form of ethnicised and securitised political order. The party-state positions its development and education strategies within existing ethnicised discourses on modernity and backwardness. The texts analysed in chapter 2 and their particular understandings of what it means to be Hanzu (modern, active, and national) and Shaoshu Minzu (backward, passive, and ethnic) in a framework of scientific development and cultural evolution are central to the education process. The production of identification and the securitisation of alternatives as ‘threats’ are the basis of the party’s drive for all education in Xinjiang to use standardised Mandarin Chinese as the language of

76 Chapter 4 will return to the theme of medicalisation of security threats in official and popular Han discourses on Uyghur-ness.

125 instruction since 2004. As the Ethnic Minority Policy text explains, education using Mandarin Chinese goes hand in hand with the transmission of “minzu tuanjie education”, of which The 50 Whys and its nationalist discourse are a central component (State Council, 2009a, p.40). Compulsory education using Chinese as the language of instruction then has been employed as a political tool of nation-building to produce students who identify themselves as Chinese nationals through the party-state’s securitised representations of what it means to be Chinese.

Compulsory Education: Practical Measures (2001), article 15 repeats the now familiar relationship between identity and modernity. It states that teachers must “deeply love the cause of Chinese socialist education” and “strive to raise their ideological and cultural levels”. Article 17 explains that as part of the teacher’s qualification system, the regional government issues certificates upon successful examination to teachers who have attained a high level of “civilisation” and the “adjustment” of roles for those who fail. This has made political examinations, including ethnic unity education, a legal requirement for teachers of all kinds in the region. This cannot guarantee self-identification with the party and China. However, it does ensure that qualifications for teachers are such that they must be able to publically represent the party’s representations of history and security-identity. This ensures that the Chinese nation is articulated to all students in Xinjiang through the mutually reinforcing dichotomies of Hanzu/Shaoshu Minzu, security/danger and modern/backward.

The party-state’s “securitisation” of identity presents what appears to be an internal security dilemma between competing ethnic groups as theorised by Paul Roe (1999) and applied to Xinjiang by Michael Clarke (2007b). However, it is better to view contingent political and social responses in terms of process and ethnicisation rather than rigidity and fixed ethnicity. The consequences of the party’s “securitisation” of identity can be understood in terms of Frederick Barth’s (1969) conceptualisation of ethnicity as the “political organisation of cultural difference”. The party is spatially and socially ordering Uyghurs and Xinjiang as part of the Chinese nation but on a lower rung than the imagined position of Han Chinese-ness. However, the subsequent unintended consequence of Uyghur counter-securitisation of their own ethnic identity as we saw in Wild Pigeon is a rejection of this spatial ordering with the Han at the centre. Furthermore, it is unhelpful to theoretically strip this security dilemma down into internal or external boundaries because these boundaries are mutually constitutive of one another. The boundaries of Uyghur-ness, as a social category of self-identification, cross national boundaries into Central Asia through connections of language, religion, and trade. As a political category, Uyghur-ness cannot be understood as outside of Chinese politics for its peripheralisation plays such a central role in discourses of Chinese-ness. However, Uyghur-ness can be solely understood within the territorial framing offered by China. It is frequently linked to both a constitutive outside and a physical outside through discourses of “terrorism” linked to Islam and

126 the West. Xinjiang is thus an internal/external security dilemma for the Chinese state because. Xinjiang is performed through the domestic and the international in ways which make the internal and the external indivisible.

Conclusions

East Turkestan, as a discourse, is central to Chinese nation-building in Xinjiang. The identification of Turkic ethnic enemies within and supported by terrorists and “the West” from outside is a discourse of identity-security, which mutually constitutes a unified national Chinese self. This chapter adapted the concept of securitisation to show that it is a political act of boundary production rather than a reflection of pre-existing, objectively verifiable identities. It showed how securitisation is central to the CCP’s nation-building project in Xinjiang because it demarcates the boundaries of the nation by articulating who and what is a threat to its existence. The first section showed how the official discourse of East Turkestan designates Turkic-ness as an outside danger to produce a relational inside of China (Zhonghua Minzu) as secure. Following Butler (1999) and Winichakul (1994), we saw how this exclusionary categorisation relationally produces nationhood by negatively elaborating a “constitutive outside”. This discourse was shown to articulate internal/external boundaries, whereby the external physical borders and internal ethnic boundaries cross-over and mutually reinforce one another. The second section examined the party-state’s explanation of incidents of unrest and dissent in Xinjiang through the discourse of “The Three Evils”. This showed how all resistance to the identity offered through Zhonghua Minzu is securitised as terrorism and criminalised as ethnic separatism. However, this securitisation offers a vocabulary of resistance to nation- building as Uyghurs identify with a glorious Turkic past as a means to reject China’s 5,000 years of minzu tuanjie. The ongoing articulation of danger, backwardness, and outside-ness attributed to the category of Turk is being (re)performed by Uyghurs to re-assert the importance to Uyghur identity of the characteristics deemed unbefitting of Chinese-ness (language and religion). The final section explored the indivisibility of the internal and external boundary through the party-state’s spatial boundary drawing within Xinjiang between more and less secure and more and less modern regions. This spatial boundary was frequently linked to both a constitutive outside. However, it was also projected onto a territorial outside through discourses of the threat of terrorism linked to Islam and “the West”.

State-centric discourses of security and sovereignty dominate our thinking of who we are by drawing the boundaries of who and what is to be secured: us versus them (Walker, 1990, p.6 & 13). However, the case of Xinjiang shows how we need to look beyond the boundaries of states to see how the articulation of Self and Other though security discourses cannot be discretely

127 divided by borders but exist inside, outside, and in an ambiguous position between the two. State-centric security discourses, like all discourses on community boundaries, include and exclude and in ways which link unbefitting characteristics internally to external categories. The case of China in Xinjiang and Xinjiang in China shows us that security discourses like all identity discourses tell us who we must not be (Turks) and who we must become (Zhonghua Minzu). Xinjiang’s relationship with the Islamic and Turkic-speaking world is securitised as a threat to the identity-security of China. These exclusions produce the nation by demanding identification with the state and its model of identity as the guarantor against the threat of anarchy and danger from the outside (Campbell, 1998, p.12). These exclusions are central performative acts of nation-building in Xinjiang through which the internal/external boundaries of China are articulated. Zhonghua Minzu is thus relationally performed as non-Turkic Hua culture and non- Islamic atheism. This Han-centric model of nation-building converts rather than simply includes the non-Hua Uyghurs. However, by incorporating these ‘domestic’ discourses into the global “war on terror” and linking Uyghurs to the global threat of “Islamic terrorism”, Uyghurs are being represented as a threat to global security. China is thus being relationally represented as a source of global security and peace protecting the world from this threat of Uyghur “terrorists”. Xinjiang is best framed as an internal/external security dilemma for the Chinese state. It presents a domestic and an international dilemma in ways which challenge what it means to be Chinese and which make the internal and the external indivisible. The next chapter will explore how the boundaries of inclusion/exclusion articulated within these security discourses were performed in the politics of the everyday after the violence of July 2009.

128

Chapter 4: Performing the Securitisation of Uyghur Ethnicity

Introduction

The previous chapter showed how the concept of securitisation can be used to analyse political acts of boundary production. Securitisation is central to nation-building because it demarcates the boundaries of the nation by identifying threats ‘outside’ these boundaries. The discourse of Turkic ethnic enemies within supported from the outside by terrorists and “the West” relationally constitutes a unified national Chinese self. Taking the nation as an un-problematised unit of analysis obscures the relations of antagonism between different social groups in interaction with the discourse of the state. Reifying the ethnic and national boundaries of the units of analysis we seek to explain would hinder understanding of identity as a process as well as exclude groups who contest these boundaries. And as scholars of Chinese Studies, this means problematising popular conceptions of violence in Xinjiang as intrinsically ‘ethnic’ and caused by a conflict between essentialised ‘Han’ identities versus ‘Uyghur’ identities. The international media used key phrases such as “race riots” to describe the violence of July 2009 (Moore, 2009). Ethnicity is an important dimension of violence in Xinjiang but framing violence as “ethnic conflict” suggests ethnicity is a cause of violence. Ethnicity and minzu are disputed categories and ethnic groups are sites of internal contestation. This follows Stevan Harrell’s understanding of “ethnic groups in action”, which are defined with reference to external relationships with other groups and the state (Harrell, 1990, p.516). It is thus important to use but look beyond official sources to understand the political imperatives through which violence is represented. Ordinary citizens have less input into official security discourses than state leaders but official and unofficial performances shape and respond to each other through constitutive and often antagonistic relations. National identities and how they are to be secured are matters of ongoing production and performance at the level of states as much as that of the household.

This chapter will explore how the boundaries articulated within identity-security discourses were performed and (re)performed in the politics of the everyday following the violent events of July 2009. On the surface of the texts of the party-state, nation-building appears as a singular, tightly controlled discourse. However, nation-building discourses are neither singular nor controllable once their ideas are publicly disseminated and employed by different individuals for different ends. The grammar of legitimation used by the state will be used as resistance against the state (Shue, 2004, p.24). The analysis in this chapter will focus on the conflicting conceptions of nation that power and resistance produce in a process of mutually shaped and

129 dialogical antagonisms. The chapter will explore how the antagonisms between the official and the unofficial are popularly challenged by different social groups in different ways, antagonising each another and thus shaping the response of the state. The analysis in this chapter is about the politics of the everyday and sits between the “textual attitude” used in chapters 2 and 3 and the ethnographic interview materials analysed in chapters 6 and 7. Section one will analyse the preceding events and the gendered and ethnicised representations of the Shaoguan incident in which two Uyghurs were killed and hundreds injured. The section will use public media, official sources, and interviews with witnesses to explore how these events shaped those of July 2009.

Section two will explore how official explanations of violence between groups within Xinjiang securitise the boundaries of national identity. It will analyse how similar acts of violence by Uyghurs and Han were dichotomously framed as threats linked to outside danger or as referents of security respectively. This section will explore how different levels of securitisation for different acts of violence by different groups can produce different subject positions within the nation. The final section will analyse the aftermath of the violence of July and its representation through the mutually constitutive dichotomies of security/danger, inside/outside, and ethnic/national. It will explore how these dichotomies continued to be performed official politics through the representation of small-scale syringe assaults in the city in the first sub- section and through popular Han Chinese protests for increased security measures in the second. Many of the sources used here are local newspapers and press conferences. These were the sources of information that Ürümchi residents had access to at the time given the closure of the internet and international communications. The analytical concern is how official discourse shaped and was shaped by the unofficial politics of the everyday. State and society in Xinjiang are best understood through mutually responsive relations. The argument presented in this section is that the national identity offered by the party-state does not cause popular Han nationalism in the region but that popular and state nationalisms draw from each other in their articulation of identities and social demands.

4.1: Producing Meanings of Violence: Securitisation and the Internal/External Security Dilemma

This section will analyse how the violence of June 2009 in Guangdong shaped the events of July 2009 in Ürümchi but were represented in different ways through discourses of danger. The analysis will explore how historical and cultural political discourses were applied to the contemporary empirical cases of Shaoguan and Ürümchi 2009 and how this shapes the governance of people in everyday settings.

130

There had been no reported incidents of violence attributed to “The Three Evils” between 1997 and 2008, so in some ways, this violence represented a “departure from, not a culmination of, past patterns” (Millward, 2009b, p.348-349). However, there had been indications of growing unrest in the region for some time. Uyghur journalist, Heyrat Niyaz, had warned the regional government days before the incident. Niyaz explained that “bilingual education” and the government policy of sending Uyghurs to other parts of China to work had angered nationalists as well as concerned professionals whose jobs were under threat (Asia Weekly, 2009). Sean Roberts expressed surprise that such tensions “had not boiled over into such violence until now” given the historically recurring independence movements and Uyghur resistance to China since the incorporation into the Qing Empire in 1759 (Roberts, 2009b). While no one predicted the nature of the violence directed at ordinary Han Chinese residents, Uyghur dissatisfaction has been evident for some time. What was new about this violence is that it was targeted at ordinary people due to their ethnicity. As nation-building accelerates its attempts convert ordinary people, through the “people-centred” approach discussed in chapter 2, Uyghurs are turning to “people-centred” resistance. These considerations make the study of the politics of the everyday, “infrapolitics” (Scott, 1985), and “everyday resistance” (Bovingdon, 2002) all the more pertinent.

The trigger of the violence of July 2009 was the murder of several Uyghur labourers by Han co- workers in the Xuri Toy Factory in Shaoguan, Guangdong on the 26th June77. 818 Uyghurs from Shufu district in Kashgar had been transferred to the factory as part of the party-state’s labour surplus export programme (Hess, 2009). This programme was enacted in 2002 and has seen 96,000 workers from southern Xinjiang transferred to other parts of China in the first half of 2009 alone (Hess, 2009, p.405). The stated purpose of the labour relocation, which went hand in hand with 300-400 million RMB of investment from the regional government, was to allow Uyghurs to “access new ideas”, “become more open-minded”, and “access modernisation” by being in contact with other ethnic groups (Global Times, 2009). Uyghurs are transferred to regions with majority Han populations. Therefore, “other ethnic groups” can be assumed to mean Han as a source of ‘modernisation’ of Uyghurs as discussed in the previous chapter. In an editorial for the China Daily at the time, Feng Jianyong of the Chinese Academy of Sciences wrote that eastern and southern China had supported “underdeveloped” areas such as Xinjiang (Feng and Aney, 2009). This “modernisation” will then enable Uyghurs to “identify with the entire nation”, which will transform “traditionally compact ethnic communities and lead to profound changes in the ethnic relations in China” (Feng and Aney, 2009).

Xinjiang, a region with the third lowest population density in China and a net labour importer was transferring labour to regions with the highest population density. Framing these transfers

77 For example, see Hess (2009); Millward (2009b); Smith Finley (2011); Roberts (2012).

131 of labour as ‘modernisation’ conceals the nation-building imperatives behind these social relations, which are far from ‘pragmatic’. These transfers had already raised concerns amongst Uyghurs that this was “forcible relocation”, thus creating ethnicised tensions within Xinjiang (Hess, 2009, p.405-407). Hess (2009) claimed that officials specifically targeted vulnerable groups for relocation, specifically women aged between 16 and 25. One elderly interviewee confided in me that he was given no choice by a party cadre but to send his daughter to inner China to work for a manufacturing firm. Another young man expressed his anger that many families from his village in southern Xinjiang had done the same. A statement by China Labor Watch (2009) on the 7th July attempted to ease these tensions. It stated that “most workers go willingly” and recommended that Uyghurs be allowed to “leave the factory at their discretion”. However, the fact that any labourers left Xinjiang unwillingly and were subsequently prohibited from leaving their place of work will exacerbate social tensions.

Ethnicity is being ordered through the coercive organisation of positions in the labour market. These ethnicised practices fit the pattern of investment in manufacturing in South-East and by transnational corporations found by Elias (2005), including China’s special economic zones, reviewed by Penny Griffin (2007). Through labour exploitation and assignment to tasks deemed to be gender and ethnicity-specific, employment practices produce rather than reflect subject positions of class, gender, and ethnicity (Elias, 2005, p.90; Griffin, 2007, p.722). In the case of the relationship between China and Xinjiang, the roles of the state and the market are often interchangeable and converge to produce the ethnicised and marginalised positions of Xinjiang and Uyghurs in China. The owner of the Xuri toy factory, Francis Choi, is a billionaire and a member of the Chinese People’s Political Consultative Conference whose connections in government allowed him to avoid repercussions after one of his sub-contracted factories was discovered to be exporting toys painted with lead paint in 2007 (Millward, 2010, p.349-350). This is but an example of how state and market are often mutually reinforcing mechanisms in contemporary China. Overlapping social responsibilities and crossover of staff blur the lines between state and market. The state and the market are then utilised by each other in promoting social policy thus making it impossible to study market mechanisms outside of their political context.

Deng Xiaoping’s statement on ‘How to Restore Agricultural Production’ in 1962 that “it doesn’t matter if a cat is black or white, so long as it catches mice” became a catch-phrase in Chinese media to define the nature of post-1978 economic reform as pragmatic and non-ideological (People’s Daily, 2008) 78. Popular international media such as The Economist (2008) framed economic reform simply that “Maoist dogma was out and was in”. Mainstream IR

78 Another similar political slogan drawn from Deng’s speeches is “Seek truth from facts” (shishi qiu shi), emphasising ‘objectivity’ over ideology.

132 literature such as Francis Fukuyama’s ‘The End of History’ (1989) argued that the “new technocratic elite now governing China knows Even more specialised works such as Barry Naughton’s Growing Out of the Plan suggested that “reforms succeeded because Chinese leaders were flexible and pragmatic” (Naughton, 1995, p.22). Powerful political institutions followed suit, for example, the US Department of State (2011) explains that Deng “adopted pragmatic economic policies at odds with Mao's revolutionary vision”. These official Chinese and mainstream IR approaches frame the particular policies of as pragmatic and technical. This claim of “pragmatism” assumes “free-market” economic practices and policies such as the promotion of exports, foreign investment, and entry into the WTO in 2001 to be non-political, non-ideological, and emerging from a set of objective economic “facts”.

This separation of development from ideology and economics from politics obscures the social practice and political choices inherent in configuring state-market relationships and the selection of development strategies79. Economic practices and the relationship between state and market are embedded in social relations (Polanyi, 2010). The boundaries of economies, like nations, are imagined and the embedded spatialisations in discourses of development can produce cognitive spatialisations which render it ‘inevitable’ through policies of technique and adaption (Cameron and Palan, 1999, p.269-270). Who are the subjects of development and who is to be secured are intimately related theoretical commitments through which states articulate nationhood and its boundaries. There is nothing natural, given, or pragmatic about economic practice, it is produced by and reproduces social relations based on identity-security discourses.

Development policies in China then can be analysed in terms of their ethnicising and gendered effects. These effects can be seen in how the events leading to the violence of July 2009 unfolded. On Friday 25th June 2009 Han workers entered the dormitory of a number of Uyghur workers and attacked them with knives and metal pipes leading to a large skirmish between both groups, which left two Uyghurs dead and hundreds injured (Hess, 2010, p.404). This violence occurred after a post on a local website said “Six Xinjiang boys raped two innocent girls at the Xuri toy factory”. The China Daily originally reported that the incident “had been triggered by a sex assault by a Uyghur worker toward a Han female worker” (Smith Finley, 2011, p.90). These allegations later turned out to be a fabrication by Zhu Gangyuan, a worker who failed to be re-employed at the factory after quitting his job (Xinhua, 2009p). Nevertheless, claimed that 20 local interviewees, including the police, believed the allegations and that published casualties figures were kept artificially low (Watts, 2009). One young man said “I just wanted to beat them. I hate Xinjiang people…we used iron bars to beat them to death” (Watts, 2009). One local shop owner drew from the party-state’s discourses on nation-

79 For examples of critical IPE, see Susan Strange (1970), Robert Cox (1981), Cameron & Palan (1999), and Nicola Phillips (2005).

133 building and ‘modernisation’, saying that “Xinjiang people have a low level of civilisation…they chased and harassed the girls all the time”. These stereotypes are common in Xinjiang where Han frequently represent Uyghur men as “sexual predators” (Smith Finley, 2011, p.82). Uyghur masculinity is then represented as a threat to Chinese identity-security. These stereotypes draw on official discourses of Xinjiang as “backward” and “dangerous” but also as a masculine threat which is not civilised enough to contain its own burgeoning and primordial sexual instincts. Sheehan (2009) and Millward (2010) both comment on alternative stories appearing in foreign media where a Han Chinese woman entered the Uyghur men’s dormitory by mistake. The men then shouted “boo” and her surprised scream led to the story that she had been raped. However, both versions are largely the same narrative of dangerous, backward Uyghur men committing sexual violence against and thus producing vulnerable Chinese women.

The Shaoguan violence was gendered and ethnicised at the same time through anxieties that Uyghur men represented as “sexual predators” are a threat to Chinese identity-security 80 . Chinese representation of minzu in both popular media and academic discourses as exotic, erotic, and primitive plays a pivotal role in the construction and essentialisation of the majority Han identity as both normal and civilised (Gladney, 2004, p.13-16). Minorities are represented as agricultural, feminine, out of step with modernity but as integral components of Zhonghua Minzu (Schein, 2000, p.130). Shaoshu minzu are incorporated into Zhonghua Minzu are included into China through representations of passivity, femininity and performances of happy dancing. Uyghur men thus present an anomaly within this discourse because they are represented as a masculine threat to identity-security. James Millward was correct to attribute the incident in part to what he called “gender anxiety” (Millward, 2010, p.350). However, the violence reflects anxiety about gender and national security at the same time. Gail Hershatter (1993; 2007) showed how representations of Chinese women during the Republican era were important concerns of reformist intellectuals charting the “fate of the nation”. Transgression of ethnic boundaries is often represented in terms of protection or domination of the female body (Smith Finley, 2011, p.82). In this way the alleged rape by Uyghur represented an attack on China and a refutation of the position of Uyghurs as passive and feminine as much as an attack on an individual woman.

Videos of the Shaoguan incident were quickly posted on YouTube (2009) and Chinese video hosting sites, which went viral across Xinjiang. The guilty parties received no punishment at the time and the absence of action sparked protests discussed in the next section81. It was October 10th before two Han men were sentenced (Smith Finley, 2011, p.75). The Shaoguan and

80 I explore the relationship between discourses on gender and ethnicity in Xinjiang through representations of Rebiya Qadir further in ‘Gendering Ethnicity, Ethnicising Gender: On the Securitisation of Uyghur Feminity’ (forthcoming). 81 Several months later the perpetrators were given unspecified punishments announced not only on the news but sent as SMS messages by the public security department to every mobile phone in Xinjiang.

134

Ürümchi violence were handled differently such that the former saw no security crackdown or news coverage until the incident was spread across the blogosphere (Smith Finley, 2011, p.78). The violence against Uyghurs in Shaoguan was represented as an “ordinary public order incident” whereas rioters in Ürümchi were labelled as the “scum of the nation” (Ministry of Information, 2009, p.106). Shaoguan government spokesman, Wang Qinxin, said the factory killings were a “very ordinary incident”, which were exaggerated to foment unrest (Watts, 2009). Development and hierarchical ethnic relations were mutually productive within mutually reinforcing discourses of modernisation, ethnicity, and security. One policeman asked the assembled journalists in Ürümchi “Why are you reporting on Uyghurs? The Uyghurs chopped the head off 100 Han Chinese. Why don’t you report that?” (Foster, 2009). This exemplified how even people in official positions people were ethnicising violence. “The Uyghurs” are thought of as “scum” who do not deserve media attention unless it is framed through the repetition of ethnocentric stereotypes.

The protest for action against the murders of the Uyghurs in Guangdong had been organised on the internet in the early days of July. It began with a march at 5am (Xinjiang time) on the 5th July from the South Gate (Nanmen) to the People’s Square (Renmin Guangchang) in the city centre of Ürümchi and involved in the region of one thousand Uyghurs, including those heading the march with large Chinese flags 82 and chanting “justice” and “equality” (Millward, 2009, p.351-352; AI, 2010). As Chapter 3 discussed, most Uyghur protest has centred on claims of inequality in the PRC and to cultural distinctions against the Han. Here, the protest was against injustice and inequality whereby Uyghurs could be killed by Han but no prosecutions appeared imminent. Without interviewing organisers, the conclusions we can draw are limited. However, we can see how Uyghur resistance was taking place within discourses of power and of China. Here we see a performative enactment of plurality and equality where Uyghurs were demanding the inclusion, acceptance, and freedom of assembly in a multi-ethnic China, which they currently do not enjoy. Like Judith Butler’s celebration of the American being sung in Spanish, these (re)performances were an “articulation of plurality” which ought to exist (Butler and Spivak, 2007, p.60-61). However, the official Chinese accounts only mention this protest to dismiss it as “lies” from Rebiya Qadir and “The Three Evils” to stir up separatism (Ministry of Information, 2009, p.17). This silenced Uyghur identities from the public record. The protest was not simply an illegal interference with public order as all protests in China are labelled. It was represented as an existential threat to a unified and harmonious Zhonghua Minzu because it challenged the idea that Uyghurs are satisfied with their social position in China.

82 Several videos confirming these events were posted on Youtube at the time but have since been removed.

135

According to official Chinese government figures the violence of July the 5th 2009 in the capital city of Xinjiang, Ürümchi resulted in an official reported 197 deaths, over 1,721 injuries, the “looting and burning” of 131 shops, 633 “damaged” houses, and 1206 destroyed vehicles (Ministry of Information, 2009, p.3-4; Xinhua, 2009c; 2009d). The mixed Uyghur-Han area of Zhongwan Street, south-east of Yan’an Lu saw Han shops torched as well as people dragged from their cars and beaten (Millward, 2009, p.352). Uyghur, Han, Hui, and foreign eyewitnesses told stories of Han being thrown to the ground and stamped to death by crowds of Uyghurs, car-jackings, and buildings being set ablaze83. Impromptu roadblocks were then built by rioters in Saimachang, the Xinjiang University area, parts of Yan’an Lu, and the Erdaoqiao/ Döngköwrük shopping area. A number of informants of various ethnicities had to hide in Xinjiang University when the violence broke out. One foreign eyewitness left the main buildings on the morning of the 6th after People’s Armed Police had secured the area and a clean-up operation had taken place. However, they explained the streets were still stained with blood, which was running into gutters, and there remained intestines and body parts strewn around Victory Street (Shengli Lu). The violence was abhorrent but should be understood within the politics of the region. Nation-building has enabled demographic transformation of Xinjiang under the justification of modernisation and nation-building. It is no surprise then that the officially articulated sources of this “modernisation” (i.e. Han people) were targeted in acts of resistance. Many such targets were also symbols of ethnic inequality such as car dealerships, symbols of luxury but which tend to be owned by Han (Smith Finley, 2011, p.79-80). Rioters attacked the perceived threat to their identity-security: an essentialised Han Other understood as the cause of economic inequality and the marginalised social position of Uyghurs in China. People-centred nation-building was producing violent resistance against people as symbols of this project.

Videos of men, women, and children being killed by various means were being passed around on mobile phones and initially were placed on Chinese video hosting sites such as Tudou and Youku but were removed shortly after. The violence touched the whole city and every single eye-witness I spoke to was clear that this was ethnically targeted violence from both Uyghurs and Han84. Multiple Uyghur, Han, Mongolian, Kazakh, Hui, and foreign eyewitnesses told me they were able to placate different groups of rioters and evade violence by declaring their ethnicity. The same stories were retold to me by Uyghurs, Han, and foreigners about attacks on Uyghurs and Uyghur-owned businesses in the Han-populated north and centre of the city on

83 Interviewing on this topic was obviously problematic given the trauma people had experienced. Uyghurs and Han alike were very eager to tell me their stories of why they thought these events occurred, which is discussed more fully in the final chapter. Most people were open to explaining where and when events occurred. Unsurprisingly, they were less open on discussing the details of the violence. I chose not to press people for further details because it was their choice how they wish to discuss or not discuss their trauma. 84 For examples of international media coverage and other eyewitness accounts, see (2009); Millward (2009); Smith Finley (2011).

136 the 7th July 85 . Large numbers of Han residents organised themselves into mobs to enact revenge attacks on Uyghurs across the city. There is no official information available on these events but there are no recorded arrests. The targeting of businesses and residents by Uyghur rioters due to their association with Han ethnicity created insecurity for Han as a group. More than a dozen local Han informants explained the officially unreported retaliatory ethnically targeted violence against Uyghurs after “7-5” such that the government was “too slow” and “too soft” in dealing with “Uyghurs”, “terrorists”, and “East Turkestan” (dongtu)86.

All Han residents I spoke to were well aware of the violence against Uyghurs on the 7th. Most interviewees were eager to tell me the “truth” about “7-5” with more than a dozen interviewees expressed pride regarding the events of “7-7”87, saying “we battered the Uyghurs!” or “we proved we too are a unified minzu”. Ordinary residents, seemingly outside politics, were taking the party-state’s discourse of “The Three Evils” and using it against its designated threats (Uyghurs). However, they also turned these discourses against the referents of security (the party-state), which was now being challenged as holding a legitimate monopoly on the provision of security. Peter Hays Gries argued that the party-state is losing control over nationalist discourse, citing examples such as nationalist literature which critiques party policy for accommodating the US and Japan (Gries, 2004, p.180). The argument here is that state and society are best understood through their mutually constitutive relations, whereby they each shape, reinforce, and respond to one another.

The violence made different groups feel insecure for different reasons. The violence of the 5th, which emerged from threats to Uyghur identity-security, sparked the retaliatory violence of the 7th. All non-Han interviewees explained they were too frightened to leave their homes for days because of the mobs of Han with weapons and all the screams they could hear. However, even on the 5th several Uyghur friends had to hide in buildings in Erdaoqiao overnight to hide from the continuous gunshots of the security forces. These gunshots were not understood as a means to their own security provided by the state. This violence was seen as a threat to their lives because they believed they would be targeted for being Uyghur. Uyghurs were framing the state security apparatus as a threat. The promotion of security, infused with the inclusionary/exclusionary discourses of nation-building was producing insecurity as it became understood by Uyghurs as ethnic targeting. Explaining this as an “internal security dilemma” between competing ethnic groups following Roe (1999) and Clarke (2007b) would ignore the

85 This is also supported in numerous interviews with eyewitnesses by Amnesty International (AI, 2009, p.17-19). 86 This is more fully explored in the final chapters based on interview materials. 87 The phrase “7-7” was never used in any official accounts or public media but I used it in interviews and conversations frequently with Han. They knew what I was referring to and would then use it themselves. This is an example of the awareness of these events but that people were choosing not to highlight them as they were with “7-5”.

137 role of the party-state in ethnicising politics. This was a cycle of insecurity where party-state in co-ordination with the ethnicised labour relations of the ‘market’ had sparked antagonisms between ethnic groups. As we shall see in the next section, these antagonisms provoked different security responses by the state, which created further insecurity for Uyghurs as a group. The insecurity felt by Uyghurs waiting for a response to the Shaoguan incident had sparked violence against Han as an ethnic group and a cycle of ethnicised violence emerged.

4.2: Producing Meanings of Violence: Nation-Building and the Securitisation of Minzu Boundaries

This section explores how official explanations of violence between groups within Xinjiang securitise the boundaries of national identity. It will build on the analysis in the previous section which showed how the violences of Shaoguan and Ürümchi were treated in different ways. This section will argue that how different violences by different groups in the same place at the same time can be represented in ways which hierarchically order security threats and thus the nation. Reports in the early 1990s made obscure references to protests by dates and numbers so as to “convey meaning to those in the know and remain mysterious to others” (Bovingdon, 2010, p.120). However, the violence of July 2009 became referred to as “7-5” in official documents, study-books, the media, and in everyday conversation. This has the effect of reifying the event by excluding and concealing alternative interpretations. President Hu Jintao cut short his trip to Italy on the 7th July and returned to Beijing. Cancelling his attendance at the G8 meeting indicated the seriousness with which the party-state viewed the development of events in Xinjiang and their significance to Chinese politics (Xinhua, 2009k). By the 9th July President Hu had already publicly declared that “7-5” was a grave, violent incident with a “deep political background” and part of a plot by “The inside/outside Three Evils” to destroy minzu tuanjie, harm prosperity, and split the motherland (Hu, 2009, p.1-2). The production of “7-5” served to produce a singular event on a singular date. This ‘event’ was represented as a threat to the Chinese nation, thereby excluding all violence against Uyghurs on the 7th July from this narrative. This was not solely repressive of alternatives but productive of a narrative which became take for granted reality. Different violences were imbued with different meanings in ways which ethnicise and dichotomously frame these violences as inside or outside the nation.

The WUC claimed the casualty figures were too low and over 600 people had died (BBC, 2009). The figures were also constantly disputed in interviews with Han and Uyghur locals who all believed many more had died. Some said this was to avoid more ethnic tensions. Several Han Chinese residents explained the casualty figures were kept low because if they had exceeded

138

200, then the UN would have to come to Ürümchi and “interfere with China”. State Media was keen to emphasise that the “international community” (namely Belarus, Thailand, Cambodia, and Laos) supported China’s efforts and shared its view that this was an “internal affair” (Xinhua, 2009e). For the party-state and some Han Chinese residents in Ürümchi, this was an “internal” issue and interference from the ‘outside’ was securitised as a threat. However, for Uyghur campaign groups this was an opportunity to invite the ‘outside’ to look in by publicising the violence and speaking to international news agencies. Inside and outside and security and danger meant different things for different groups, where for Uyghurs security was being sought ‘outside’ the boundaries of China, which the party-state seeks to enclose.

Official reports88 state the initial outbreak of violence on the 5th was at the People’s Hospital on Shengli Lu (Victory Street) adjacent to Xinjiang University in the south of the city, which is about half an hour bus ride from Renmin Guangchang (People’s Square). Official claims that organisation of the protest at the People’s Square was linked to the violence and that there was crossover in participation have not been corroborated with any evidence. The protest and the violence certainly started separately as the violence broke out while the protest was still happening. The 50 Whys, however, claimed that this was an internationally funded and synchronised terrorist attack because Uyghurs involved drive buses and trucks stacked with rocks (Ministry of Information, 2009, p.3-5). Using rocks held in trucks as weapons is not an indicator of synchronised and internationally organised terrorist attacks. These were riots in which Uyghurs, Han, and the security services were all perpetrators and victims (Roberts, 2012, p.15-16 & 23). It appears probable that these people-centred violences were planned by different people with no links to the protest. Like earlier incidents discussed in the previous chapter, this bore all the hallmarks of Bovingdon’s (2005) “snowballing”. The protests and the violence indicate “extensive Uyghur discontent” and some willingness to brave government reprisals in expressing it (Bovingdon, 2010, p.170). The representations of unrest as organised and orchestrated by a handful of “terrorists”, just as representations of incidents in the 1990s, was designed to deny that these were authentic expressions of mass sentiment (Bovingdon, 2010, p.121). Violence spread rapidly and chaotically across a city in which discontent with CCP policies is widespread and different individuals with different grievances became involved. In this case, the grievances were held by Han as well as Uyghurs and the “snowballing” was inter- ethnic and included security services as much as ordinary residents.

Most Uyghur informants and several eye-witnesses explained the outbreak of violence to me in terms of heavy handed policing preventing a peaceful protest with one claiming a policeman struck a female Uyghur at the square. Eyewitness interviews with Radio Free Asia (2009)

88 For example, see: Ministry of Information (2009).

139 claimed the protestors moved to Nanmen (South gate) after the police moved into the square and began detaining people. Least documented are the clashes with police while Millward (2009) and witness interviews with Radio Free Asia (2009) suggested Uyghurs fought with the People’s Armed Police (PAP) at the Rebiya Trade Building89. Informants told me they saw and heard constant police gunshots in the city centre and the Erdaoqiao/Döngköwrük area. One Uyghur informant told me he saw trucks of Uyghur corpses being driven away by PAP in the Samaichang area. James Millward (2009) and Radio Free Asia (2009) described similar events in the Dawan Xiang area. A Turkish journalist cried saying he could never report the disposal of the corpses by the security forces he had witnessed. There is little if any way such events can be corroborated. What we can conclude is that multiple eyewitnesses of different ethnic groups believe that security services killed Uyghurs and disposed of bodies before any investigation was possible. Even Uyghurs who were not eyewitnesses discussed these “rumours” and recounted these narratives as truth. These narratives were believed and retold because Uyghurs understand their own position in China through discourses of exclusion and insecurity where the party-state is a threat.

Regional Chairman, Nur Bekri (2009), and Party chief, Wang Lequan (2009b; 2009c), used their televised speeches on the 6th and 7th July to condemn “The Three Evils” “distortion of facts” regarding the Shaoguan incident. The World Uyghur Congress (WUC), the Uyghur rights campaign group headed by Rebiya Qadir, was eventually blamed by the party-state for the violence. The study documents used in all minzu tuanjie classes explained that Rebiya Qadir, the World Uyghur Congress and “East Turkestan”/ETIM collaborated to use representations of the Shaoguan incident to “harm minzu tuanjie” and “split the nation” (Ministry of Information, 2009, p.8 & 16). These organisations are then said to have “distorted facts” by representing this as a minzu problem when the violence had been a “model public-order incident” and nothing to do with ethnicity (Ministry of Information, 2009, p.8 & 16). This imbued different acts of violence with different meanings. Violence against Uyghurs, such as the Shaoguan incident, was represented as a standard legal matter to be dealt with by police and the courts. On the other hand, violence against Han, and indeed protests for ethnic equality, were securitised as existential threats to China and its identity-security such that the PLA and PAP were brought into Ürümchi to deal with the conflict. This ordering of violences orders the internal hierarchy of the Chinese nation in Xinjiang. Han bodies are positioned in the realm of security as metaphors for the integrity of the Chinese nation yet violence against Uyghur bodies is less of a security threat as a “model public-order incident”. As one Han party cadre remarked after a meeting on the 1997 Ürümchi bus bombing, “we always like to decide on the nature of

89 Rebiya Dasha was a plaza built in 1992 by Rebiya Qadir’s company when she was a successful businesswoman working in wholesale, retail, and later in real estate in the region and in Kazakhstan (see: Kadeer (2009) ‘Do You Hear the Rooster Crowing?’ and ‘The Wealthiest Woman in China’).

140 some event” (Bovingdon, 2002, p.64). Clearly, the “nature” of “7-5” had been decided very early on. Only now it was decided at the apex of the Chinese party-state which superimposed the friend/enemy defining discourse of global terrorism onto the internal boundaries of Han and Uyghur in Xinjiang. The “life of the nation” was being performed by restating the social hierarchy which takes Han bodies as intrinsically more valuable, more threatened, and more Chinese than Uyghur bodies.

Li Zhi, the city party secretary, met with Han rioters on the 7th. He chanted “down with Rebiya” on a megaphone, asking rioters to “unite and build a better Ürümchi” but without condemning the violence (Xinhua, 2009f). The absence of news coverage or official reference to the violence of “7-7”, let alone condemnation, created an imbalance in the meanings attached to violence dependent on one’s position in the Chinese ethnic hierarchy. Wang Lequan, the Xinjiang party chief at the time, made a televised speech on the 7th solely discussing violence against Han as a “strategy” of the “inside/outside Three Evils” to “destroy minzu tuanjie” (Wang, 2009b, p.8). In stark contrast, he then addressed Han rioters as “comrades” and that their “operations” were only “not needed” because the security apparatus was now in control (Wang, 2009b, p.8-9). This was an attempt to calm the violence by Han and a call for “clear heads” (tounao qingxi). However, this framed violence against one ethnic group, the Han, as a threat from outside to China and as a threat to China. Violence against Uyghurs, by contrast, was undesirable but was being committed from the inside, by “comrades”. The Han were being presented as the referent and even a potential source of inside-ness and security where “the inside/outside Three Evils”, in other words Uyghurs, were the threat. This ordered and ethnicised the meanings of violence against ordinary people such that it was being committed by one group against the nation and the other for the nation. Visceral acts of violence were being mediated through meanings attributed to different ethnic groups which performed the boundaries of friends/enemies and inside/outside in a process of hierarchical nation-building.

“7-5” and the struggle against the “inside/outside Three Evils” became represented through the constant repetition of the slogan that it was “not a minzu problem, not a religious problem but a political problem of defending the unity of the nation and the fundamental interests of the masses” (Ministry of Information, 2009, p.4 & 13-15; Bekri, 2009a; 2009b; Wang, 2009b; 2009c). On the 10th July, the casualty figures were broken down according to ethnicity and then updated in August by the state-media network, Xinhua, which stated that of the 197 killed, 134 were Han Chinese and 51 were Uyghurs, though of the latter, only 10 were “innocent civilians” (Xinhua, 2009c; 2009d; China Digital Times, 2009). The accuracy of these figures is impossible to ascertain but this did symbolically divide the casualties into almost the exact proportions of population by ethnicity of the city. The party-state’s minzu tuanjie education programmes continued to state that the casualties of “7-5” were “not only Han” but “all minzu” (Ministry of

141

Information, 2009, p.14). This enabled the ongoing representation of the violence as predominantly against Han by “the inside/outside Three Evils” 90 , which posed an internal/external threat to the entire nation. This articulated the security threat of Uyghur-ness to China by attributing all the violence to “The Three Evils”, intimately linked to Uyghur ethnicity. Linking all the casualties to the 5th of July removed the violence committed by Han against Uyghurs from the historical record. All casualties became synonymous with the struggle between the mutually constitutive unified Chinese nation and “The Three Evils”. The violence of the Shaoguan incident and “7-5” were dismissed as being unrelated to ethnic relations. This was because the participants were criminals who “cannot represent every member of their minzu” and it did not cause “” (Ministry of Information, 2009, p.14). These representations seek to silence and dismiss claims of ethnic discrimination (“equality” and “justice”) as security threats. It also concealed the ethnicised representations of violence and the ethnic targeting in the subsequent crackdown.

The party-state’s response to the violence of July 2009 targeted Uyghurs as potential enemies and an object for surveillance. In 2009, documented the case of 43 Uyghurs who have been “disappeared” since being forcibly detained by PAP (HRW, 2009, p.21- 32). Many of these “disappearances” took place during the large scale security sweep in Saimachang and Erdaoqiao/Döngköwrük on July the 6th. Uyghur groups in exile claim more than 10,000 Uyghurs are still reported missing (RFA, 2012b). One such case is Patigul Eli, who remains under permanent police surveillance. This is to prevent her petitioning to be given information regarding her missing son, Imammemet Eli (RFA, 2012a). Imammemet, an alumnus of the University of Technology, was not accused of physical violence but is claimed to have participated in demonstrations and was detained in July 2009 (RFA, 2012a). Another such case, which more starkly indicates the ethnic targeting and arbitrary nature of detentions is that of the CCP member, Abdugheni Eziz, a former village Party secretary in Khotan who owned the Yang Guangcheng real estate company. Abdugheni volunteered to provide food and drink for the PLA and PAP in downtown Ürümchi during July 2009 but has been missing since being taken for questioning without explanation on July 31st 2009 (Radio Free Asia (2012b). 18,000 arrests were made under suspicion of “terrorism” in 2005 and this created a self-fulfilling prophecy of insecurity where Uyghurs feel threatened thus contributing to events such as the violence of July 2009 (Roberts, 2012, p.15-17). There has since been a rise in the number of violent incidents across the region targeting security services since 2009 in Kashgar, Khotan, and Yecheng. The party-state’s strategy is unlikely to promote security in

90 This phrase was in continuous use throughout the political education materials analysed here (eg Ministry of Information, 2009, p.8). These classes also utilised all the key speeches of national and regional leaders in the immediate aftermath of the violence, which repeated this discourse. For example, see Hu Jintao (2009); Wang Lequan (2009b; 2009c), and Nur Bekri (2009a; 2009b).

142 the long term as it further exacerbates ethnicised tensions where Uyghurs view the party-state as a threat in alliance with Han Chinese residents.

The propaganda drive for minzu tuanjie and the claim that ethnic relations bear no relevance to these events appears impossible to reconcile. However, the internal logic of this discourse can be illuminated by understanding power relations as productive at least as much as they are repressive. The discourse (“not a minzu problem”) became a central part of the nation-building project to define what it means to be minzu (ethnic) and what it means to be Chinese (national):

Why did the hatred of a minority of thugs cause so much damage to members of other minzu? This completely explains one problem. It is not that these people have no minzu feeling or minzu consciousness, it is just that their minzu consciousness is a narrow- minded minzu feeling and a narrow-minded minzu consciousness…they are unable to correctly distinguish the true from the false or who is friend and who is the enemy…so to overcome narrow-minded minzu feelings and consciousness we must establish the correct minzu feeling and consciousness, these are manifested in loving one’s minzu as well paying attention to the development of one’s minzu as well as loving other minzu and paying attention to their development…and never wavering in the intense struggle against “The Three Evils” and self-consciously protecting national unity and minzu tuanjie”. (Ministry of Information, 2009, p.67-68)

The friends and enemies of the revolution were the first question of importance for Mao Zedong and here we see the language of friends and enemies returning through the mutually constitutive dichotomies of true/false, narrow/broad, and inside/outside. In this narrative, distinguishing the self from the enemy and the inside from the outside is about distinguishing “correct” from “incorrect” minzu identities and Zhonghua Minzu from “The Three Evils”. This securitisation of identity is not simply about policing the external boundaries of their physical beginning and end but about fixing the internal content of what it means to be Chinese and in what ways one is permitted to be Uyghur. Turkic language and Islam are officially excluded as defining features of Uyghur-ness and their importance is attributed solely to the “manipulation” of “The Three Evils” (Ministry of Information, 2009, p. 55 & 94). Identification with the categories of Turk and Muslim are thus framed as the enemy and the security threat through which the boundaries of inclusion in China are demarcated. This discourse explicitly highlights the pursuit of the “development” of minzu as a form of conversion from “narrow-minded” ethnic identity to absorption into “national unity”. “Development” is understood as unification into the Chinese nation, as the “normal” and “correct” way for Uyghurs to understand themselves and overcome their “narrow-minded minzu feelings”.

Delegitimising and silencing Uyghur resistance is important for nation-building because the discourse of Zhonghua Minzu speaks for all ethnic groups in asserting that they want to be included in China. The idea that “naïve and excitable students” are manipulated by a minority of separatists who “recede into the background” has long been a way for the party-state to delink

143 all protest in Xinjiang from the widespread discontent it reflects (Bovingdon, 2010, p.119-120). The 50 Whys91 divides participants of the July 2009 protests and riots into the instigators (“The Three Evils” and “East Turkestan”) and the “ignorant92 students” had their “minds controlled” (kongzhi sixiang) and were fooled by the instigators’ “distortion of facts” (Ministry of Information, 2009, p.96-97 & 104-107). The party-state’s cultural governance frequently invokes “facts” in a “positivist spirit” to frame cultural politics. Its documentation of the Nanjing Massacre through photographs of mutilated bodies was used to ‘prove’ Japanese “barbarism” in order to produce standard of Chinese “civilisation” (Callahan, 2007, p.7). However, in Xinjiang, the party-state uses “facts” to target and denounce those Uyghurs inside the territory of China who understand history outside the official narrative Zhonghua Minzu. The 50 Whys is a key example of this cultural governance through “facts” which blur internal and external security threats. It continues its securitisation of the “facts” of China by saying the “reactionary ideology (of “The Three Evils” and “East Turkestan”) has been instilled into the minds of youth” in an attempt to use students to “distort Xinjiang history and interfere with minzu relations” (Ministry of Information, 2009, p.96). Discord on campuses has then become “used as a pretext by anti- Chinese international forces” to cause harm to China (Ministry of Information, 2009, p.96-97). “Infiltration” of the education system through the media and the internet has thus become an “important strategy” of the “inside/outside Three Evils” which “sows minzu discord” and “attacks the party and the country” (Ministry of Information, 2009, p.99).

Education as “an important formative component of ideology” is thus a crucial site in the “struggle” against “The Three Evils” (Ministry of Information, 2009, p.96-97 & 105). This educational frontline is to ensure that “religion” or “any form of interference” does not harm “scientific culture education” (kexue wenhua jiaoyu) (Ministry of Information, 2009, p.104). This concept of “infiltration” frames the outside as a security threat but as a non-physical threat at the level of ideas. This “struggle” is particularly acute regarding the “truth” of “7-5” such that everyone must “dry their eyes and clearly distinguish enemy from self (renqing diwo)” (Ministry of Information, 2009, p.104 & 106). This is the “theoretical frontline” discussed in chapter 2, where the party-state’s minzu theory was explicitly highlighted as central to resolving the “minzu problem” by shaping people’s self-understanding. Here the “theoretical frontline” is central to nation-building to exclude enemies (“The Three Evils”) and secure the self (“China”). Xinjiang students have a particularly important role to play in this security because “compared to ordinary young people, they need more education” to “establish the correct national outlook (guojiaguan), minzu outlook, religious outlook, historical outlook, and cultural outlook” (Ministry

91 This was also a theme in the televised speeches immediately after the violence. See Bekri (2009a) and Wang (2009b). 92 This was expressed in various ways including wuzhi (without knowledge) (Wang, 2009b, p.8), “unfamiliar with national conditions”, “without experience of the world”, and “lacking clarity” (Ministry of Information, 2009, p.97).

144 of Information, 2009, p.107). Nation-building is thus targeting self-identification through education of students as to how order their understandings of nation, ethnicity, religion, and history into the “correct” “outlook” established by the party-state.

These processes of securitising the nation takes place at the level of the everyday. “Infiltration” of Xinjiang discussed with regard to “reactionary articles and internet postings” as well as “direct phone calls to homes of staff and students” (Ministry of Information, 2009, p.99). “Performances of art, poems, and songs” as well as “academic exchange and funding” are said to promote “infiltration” by “propagating Pan-Islam, Pan-Turkism, and the distortion of Xinjiang history” (Ministry of Information, 2009, p.101). “Reactionary articles” means those which prioritise minzu, Islamic, or Turkic identities over the cultural progress associated with Zhonghua Minzu, which of course includes Wild Pigeon discussed in chapter three. This is a warning to all staff and students that alternative interpretations of history and identity will be treated as threats from the ‘outside’ and as a form of terrorism under the rubric of “The Three Evils”. “The Three Evils” are said to use “everyday discussions” of ideas such as “justice” and “equality” to “harm minzu relations” under the banner of “discussing minzu history and the contemporary minzu situation” (Ministry of Information, 2009, p.99). Activities such as “making friends, recognising each others’ hometowns, and student parties” are said to be used by “The Three Evils” to “draw people in and seduce them” into “separating Xinjiang from China” (Ministry of Information, 2009, p.99). This securitisation of the everyday makes who one talks to and what one talks about a security matter which determines the very life or death of the nation. The party-state is reminding readers of surveillance on campus to instil self-censorship amongst students and staff in their discussions of life in Xinjiang. “All teachers, students, and staff must maintain a high level of vigilance” and “protecting the unity of the nation is a political demand and historical mission of every student” (Ministry of Information, 2009, p.102 & 107). The choice of who to socialise with and how thus becomes entangled in the party-state’s teleology of cultural evolution. The “historical mission” is for all citizens to build a political community by guarding against those who wish to discuss the history or contemporary politics of their nation. These representations construct an ever-present non-physical security threat in the realm of thought and identification, where “terrorism” is less an armed movement than it is a way of viewing the world and writing or talking about it. The struggle between the enemies and the self is thus one over how to understand and think about the nation.

The universities rounded up all students who were involved in the protest march during which chants for “justice” and “equality” were prominent. After reporting to party representatives at the university, these students were confined and monitored in their dormitories. They were then “educated” by being given extra minzu tuanjie classes which used The 50 Whys as a core

145 text93. These students’ “thinking had problems” and they therefore “had to learn a lesson” (Ministry of Information, 2009, p.105). The students’ “crime” as stated to me by staff responsible for their identification was that they got “too excited” and were “shouting bad things”. No students were charged with any crimes in a court of law as the people prosecuted for violence were all from outside Ürümchi and from poor backgrounds according to official reports. Nevertheless, The 50 Whys and public speeches explicitly highlighted the involvement of students and Professors. This conflated violence, peaceful protest, and discussion of ideas such as “justice” and “equality” as security threats to China and to Chinese-ness. While these students were shown some leniency their behaviour had been elevated to a level of security and beyond politics whereby they could be judged and charged without recourse to the courts or public scrutiny.

The criminalisation and securitisation of “separatist” thought rather than “separatism” as an organised violent movement conflates Uyghur articulations of identity and discontent with violent “terrorism”. The violence committed by Uyghurs against Han challenged the notion that the Han were the superior, benevolent centre, appreciated by all minzu. In July 2009, a background of publicly displayed political slogans (kouhao) was arranged by police and public workers across seemingly every inch of the city. These political texts (figures 4.1 and 4.2) offered explanation and gave meaning to the threat of physical violence in the form of tanks, thousands of PAP and PLA armed with machine guns, attack dogs, and riot police, which patrolled every area of the city.94 The most commonly observable kouhao to appear at the time and then again during the re-imposition of martial law on September 3rd was “minzu tuanjie is prosperity, minzu separatism is disaster” (see figure 4.1). This slogan was repeated throughout minzu tuanjie education manuals and was stated to be not simply a “political slogan” but a “political, cultural, and economic lifeline” (Ministry of Information, 2009, p.6)95. The identity- security of the Chinese nation was being performed through modernity (“prosperity”) under threat from backward (“disaster”) identification with Turkic-ness (“separatism”). The dichotomous framing of identity-security was captured even more explicitly in the slogan “Strengthen Minzu Tuanjie, Strike Hard against ‘The Three Evils’” (see figure 4.2), which offered people a stark choice of inclusion as Chinese (tuanjie) or exclusion as a Turkic, Islamic threat (“The Three Evils”). Violence by Uyghurs against the Han was not simply represented as a threat to the body of individual persons but to the integrity and prosperity of the body of the nation. The widespread discontent which led to the protests and the violence challenged the idea that China consists of an active Han centre which benevolently civilises the appreciative

93 This information is drawn from an interview with staff personally responsible for their identification and detainment at one university. 94 Many of these kouhao remained for months to come and were observable throughout the fieldwork but this is also based on multiple eyewitness accounts from July the 5th. 95 Minzu Tuanjie as a performance of modernity and national strength is discussed more fully in the next chapter.

146 and passive shaoshu minzu frontier. This challenge transgresses the internal, hierarchical boundaries of Chinese-ness or what Winichakul called the geo-body or the “life of the nation” (Winichakul, 1994, p.11). Nation-building was proceeding through the representation of the violence of July 5th in such ways that the exclusionary character of the discourse was producing what it meant to be secure – modern, Chinese, and Han-centred.

Figure 4.1: A slogan erected by the city government’s planning department supervision centre: “Minzu Tuanjie shi fu, Minzu fenlie shi huo” (Minzu Tuanjie is prosperity, Minzu separatism is disaster).

147

Figure 4.2: Just one example of this slogan erected approximately every 20 yards on Tuanjie Lu (‘Tuanjie Street’), with a heavily military roadblock on every junction (see bottom right): “Jiaqiang Minzu Tuanjie, Daji ‘San gu Shili’ ” (“Strengthen Minzu Tuanjie, Strike Hard against ‘The Three Evils’ ”).

Han-centric visions of social order are not limited to the domestic level in official Chinese policy discourse. “Harmony through diversity” (He er butong) has been at the centre of Hu Jintao’s “harmonious world” foreign policy and his answer to the “clash of civilisations”. Hu’s “harmonious world” approach thus intensifies the re-inscription of difference but this is a pluralism of sovereign states rather than a plurality of publics (Kerr, 2011, p.171-172 & 175). Harmony, or harmonisation, is thus a nationalist drive to make identity congruent with the PRC’s sovereign borders. The party-state’s approach to identity-security in Xinjiang and international relations was exemplified through the complete closure of the internet and international phone calls for approximately 10 months after the violence of July “to prevent unrest” (Xinhua, 2009). The physically external world of international news is thus understood and represented by the party-state as a threat to identity-security. Uyghurs are quarantined from international news to prevent “infiltration” of alternative information from the ‘outside’. Conversely, this approach to security also quarantined the perspectives of Uyghurs from being

148 heard across the world. Many of these perspectives offer alternatives to Zhonghua Minzu and would contradict the official representation of China as a “harmonious society” promoting a “harmonious world” of “common security”, “common prosperity”, and “inclusiveness” (Hu, 2005).

In 2009, China signed multi-billion dollar contracts to extract natural resources bilaterally with Cambodia, , Thailand, and Pakistan and through the mechanism of the Shanghai Co- operation Organisation (SCO96) with Kazakhstan. These deals were all signed after these states breached international law to deport Uyghur asylum seekers, who had fled China after July 2009 (HRW, 2009b; 2011; Radio Free Asia, 2011b; UAA, 2012). These Uyghurs have since disappeared, been executed, or sentenced to prison in closed trials (HRW, 2009b; 2011; Radio Free Asia, 2011b; UAA, 2012). This party-state utilises its growing economic leverage and position as a “responsible power” at the international level to legitimise ‘domestic’ policy and its model of nation-building. The quarantine of Xinjiang and the securitisation of alternative forms of self-identification should not be seen as contradictions to Hu’s “harmonious world” and China’s opening up. Rather, these are best understood as integral components of a harmonisation process which require conversion into a “harmonious world”. It is disruption and resistance by the Other which produces the need for and meaning of harmony. The “reinscription of uniformity and difference” generates corresponding responses from both ‘domestic’ and ‘international’ frontiers (Kerr, 2011, p.167). Harmony thus produces and requires differentiation and contradiction to maintain its dynamism in articulating the inside/outside and the friends/enemies of China.

This regionally-differentiated approach to security encloses Xinjiang into China while the rest of China is “opening up”. It takes the maintenance of ‘domestic’ “harmony” as its referent. Harmony is understood as the ongoing production of China as a co-operative and friendly international power. However, this entails domestic hierarchy with a “benevolent” and modern “Han nucleus” at the centre of the nation. These exclusions of alternatives to the party-state’s model of China are inseparable from its model of inclusion into Zhonghua Minzu. Turkic history is excluded because it is not ‘Chinese’ and by including Uyghurs into ‘Chinese’ history, they are not Turkic. By framing violence against Han as a threat to the nation they are included within the boundaries of China but this takes on meaning when contrasted against the representation of violence against Uyghurs as a lesser problem. Securitisation is a discourse, which continuously performs and produces the contingent boundaries of identity as referents of security.

96 On the SCO as “regionalism with Chinese characteristics”, see Kerr, David (2011).

149

4.3: The City After “7-5”: (In)Securitisation and the Ethnic Threat

This section will analyse how representations of the violence of July 2009 continued to be performed and (re)performed long after the event through official politics and the unofficial politics of the everyday. Peter Hays Gries (2004) argued that the party-state is losing its grip over the monopoly of Chinese nationalism and its ability to articulate Chinese-ness. Communications technology has “opened up new spaces…for the articulation of alternative national imaginaries” loosening the party-state’s monopoly on meanings associated with being ‘Chinese’ (Leibold, 2010, p.24). However, the approach here is not to understand these alternatives as in a state of juxtaposition. Instead, the focus is how the official and the unofficial shape one another in a multiplicity of identity-security discourses. The approach here follows the insights from critical security studies that “security often produces insecurity” (Weber, 2011, p.1021). Security dilemmas emerge in the areas of identity-security when one identity is secured at the expense of another. In our case the party-state’s conceptualisation of Han-ness is being secured at the expense of Uyghur articulations of self-identification. The argument is that the party-state’s attempts to secure Han-ness from the threat of Uyghur-ness was highly destabilising at the level of social interaction and everyday politics. Security and insecurity are intimately tied so it is useful to reconceptualise securitisation as (in)securitisation to avoid concealing the insecurities which it produces. This section uses participant observations of (in)securitisation practices of the state and the popular responses and mediations of these practices. Achieving security may be illusive (Weber, 2011, p.1021). The analysis here centres on how discourses of security and inclusion produce security for some at the cost of insecurity and exclusion for others.

When the violence dissipated, there were (re)performances of patriotism as Han-owned shops and homes hung flags and slogans about Chinese history (see figure 4.3). Han nationalist self- identifications emerged in ethnically targeted-violence and the popular adoption of the discourse of ancient Han jurisdiction over Xinjiang. The party-state continues to shape Han nationalism through discourses of security which other Uyghur-ness such as the idea of the Han as nucleus and shaoshu minzu as “backward”. The oft seen slogan adorning military trucks stationed following the imposition of martial law in July was “Deeply love the frontier” (re ai bianjiang). This slogan is a modification of the slogan “deeply love the motherland” (reai zuguo). However, this modification marks Xinjiang off as a frontier distinct from the centre. In Chinese this “deep love” is normally reserved for aspects of the self such as family, the motherland, or during the communist era in slogans regarding one’s workplace. This love expresses a “we”- ness. By hanging it on military trucks policing the city, it articulates a conceptualisation of identity-security, which seeks to secure the region. However, it Others the distant, perpetual

150

“frontier” as outside the Han nucleus doing the nation-building. Xinjiang can become the self but it is not yet of it.

Figure 4.3: “Xinjiang has since ancient times been an indivisible part of the motherland” (Xinjiang zigu yilai jiu shi zuguo buke fenge de yi bufen). This kouhao is a standard political slogan which appeared all across the city and in state media broadcasts. However, this is an improvised banner a Han Chinese resident pasted on the front of a large apartment block in the north of the city. It remained there from July 2009 at least until the end of fieldwork in August 2010.

After the events of July 5th, martial law came into effect at 9pm on the 7th July under the rubric of “traffic controls” (jiaotong guanzhi) (Wang, 2009b, p.10). Two days later, tens of thousands of troops poured into the city from their regional headquarters in , Gansu province to supplement the police, People’s Armed Police, and Special Police who already patrolled every corner of the city and blocked all major traffic junctions. “7-5 was over”, as if it were a brief anomaly in the unfolding of the Chinese cultural teleology (Hu, 2009, p.1). When I arrived in the city on the 3rd September 2009, it was evident through conversations with local residents that fear of the “threat” presented by Uyghurs was still present. Most Han interviewees warned

151 me not to visit the “dangerous” Uyghur areas. Several told me they had not left their street since July. Within a single day of walking across the city I observed multiple instances of Han Chinese parents physically dragging their children away from Uyghurs minding their own business on the street. Ethnic tensions are not new to the city. However, the fear of Uyghur bodies was being exacerbated by the official securitisations which reminded citizens that they were responsible for maintaining security against the threat of infection by Uyghurs. Insecurity was exacerbated because (in)securitisation targeted groups and ideas about group-ness rather than individuals.

4.3.1: Syringe Assaults and Everyday (In)Security

A series of small-scale assaults and robberies across the city by Uyghurs using syringes against Han Chinese began shortly after “7-5” and continued through until October. While SMS messaging had been shut down by the regional government in the name of security, all residents with a mobile phone received regular ‘information’ texts from organs of the party- state97. For example, the Xinjiang News Centre texted residents on the 28th August with a security warning that “some people” had been attacked by syringes and that all residents should “be on their guard” 98 . The Regional Government announced that between the 20th August and 2 September there had been 476 such incidents in which 531 people were injured and that citizens should “raise their guard and increase self-defence”, a warning repeated in a press conference by the Public Security Bureau (PSB) (Chen Bao, 2009c; 2009e; 2009g). These statements articulated a social dynamic whereby security was something to be achieved by ordinary members of the nation rather than coming under the sole jurisdiction of the formal institutions responsible for national security. The Regional Government released regulations at the end of December which were studied at all work units in the city to “strike hard” against “The Three Evils” by emphasising the “role of the masses” and “the people as the frontline” in fighting terrorism and East Turkestan (XUAR, 2010, articles 5, 9, & 16). The individual citizen was bound by and responsible to uphold these regulations (XUAR, 2010, article 2). Security here was being discursively positioned as beyond politics and the nation was a taken for granted and unquestionable referent. However, security was also below politics in the sense that it was permeating as well as ethnicising the daily practices of ordinary citizens. For example, in the residential area in which I lived, non-Han friends were prevented by part-time security guards from entering our living area on several occasions while Han Chinese visitors were not subject to this surveillance.

97 This is a particularly effective means of disseminating political propaganda. According to the Ministry of Industry and Information Technology, China today has more than a billion mobile phone uses (Shanghai Daily, 2012). 98 The SMS texts referred to here were sent to every phone, including my own. Those dated before I arrived in the city were shown to me by residents.

152

By the 16th September the Public Security Bureau (PSB) had announced 75 arrests from 36 incidents across Xinjiang, 16 of which were in Ürümchi (Chen Bao, 2009g). The PSB sent a public SMS message on the 6th September stating that anyone found guilty of syringe stabbing would receive between 5-15 years in prison for “causing panic”. It was stated this would be the case even if this was without the chemicals which were rumoured to spread HIV. The Regional Public Security Department released an announcement that “no matter what tool is being used” these attacks were “creating an atmosphere of terror and interfering with social order” (XUAR Gonganting, 2009). This was less about the physical harm of these attacks but more about social order. The social order of the “new China” was threatened because ethnic majority bodies, the source of political loyalty and patriotism, were being threatened.

On the 9th September, Du Xintao, the head of the Department of Law and Regulations at the Regional Public Security Department explained in a press conference that “while the overall good direction of society is moving towards development, enemy forces were organising crimes...we must protect the masses life property security and social order” (Chen Bao, 2009i). Threats were being produced through representation where what could be a mugging was now invoked as a threat to modernity and national security. President Hu Jintao had stated that “7-5 was over” but an SMS text on the 13th September stated “syringe attacks were a continuation of 7-5” committed by terrorists, against whom all citizens had a responsibility to “resolutely struggle” with “one’s own power” for the cause of minzu tuanjie. The discourse which framed the violence as “continuation of 7-5” and committed by the “inside/outside Three Evils” against the whole nation was repeated frequently in editorials in the local newspapers, including one such statement from Zhu Hailun, secretary of the Politics and Law Committee and a member of the Regional Party Standing Committee (Chen Bao, 2009d; 2009e). The city government had placed public announcement posters across the city on the 2nd September stating that syringe attacks were “not normal public order incidents but attempts by the minority of enemies to destroy the city’s peace and unity”. The syringe assaults then became represented through the generic existential threat of the “inside/outside Three Evils”, ranging from student protests to “separatist history” to rioting and looting all to dismember the motherland. However, some of these incidents were openly explained to be muggings but they remained listed as incidents of terrorism and represented through the discourse of the enemy of the “inside/outside Three Evils” (Xinhua, 2009n). The violence was imbued with meaning by being located in the discourse of “the inside/outside Three Evils”. The narrative which took Han Chinese bodies as metaphors for the integrity of the nation was being replayed but in ways which produced insecurity. The articulated threat was not then of individual criminals to individual bodies but of one group, Uyghur terrorists, to another, the Chinese nation.

153

On the 5 September, Xinhua reported that such syringes contained drugs but China’s Academy of Military Medical Sciences announced that it would take up to 6 months of close observation to determine if victims had been infected with radiation, anthrax, hepatitis, or AIDS (Xinhua, 2009m; 2009n). The production of medicalised existential threats to identity-security had already enabled more severe punishments than would be the case had they been treated as ordinary instances of crime. The representation of these incidents as infectious threats to the national geo-body performed a terrorist Other to be excluded for the sake of the health of the Chinese Self. The geo-body of a nation occupies a geographical position but this is not merely a territorial space but the “life of a nation and source of pride, love, hatred, passion, reason, unreason” (Winichakul, 1994, p.11). The geobody, articulated through the party-state’s securitisation practices, is one which is inherently exclusionary because it seeks to secure the national majority while (in)securitising minorities.

From September 17th until the 22nd a spate of seemingly copycat syringe stabbings by Han Chinese perpetrators were reported in Xi’an yet the nationalist framing of separatist threat and national security were absent. This was presented as an ordinary case of “criminal investigation” and the Han were not being subjected to the same surveillance of identity as Uyghurs (Chen Bao, 2009n; 2009o). By mid-September 2009, Xinhua explained that the actual bodily harm committed by the syringe assaults was minimal and no victims contracted any diseases from these attacks (Chen Bao, 2009L; 2009m). On the 3rd September, the PSB texted to tell the city that only 15 people had been genuinely injured in these attacks. However, the articulation of such threats had already justified an ongoing military presence across the city, which ethnicised danger and security.

By the 4th September, there had been 476 officially reported incidents at hospitals but only 89 clear cases of syringe attacks (Xinhua, 2009L). (In)securitisation was taking place as with every securitising move people felt more insecure about the nature of the threat in their neighbourhoods. People were panicking as evidenced by the huge gap between the number of people claiming to have been attacked and those who had actually been attacked. For example, Jackie Sheehan (2009) reported that many such incidents turned out to be mosquito bites. In the city, rumours spread 99 that that there had been closer to a thousand such incidents. However, a number of residents, including one doctor, claimed patients were visiting multiple hospitals to get different opinions in the hope that they could receive compensation from the regional government. Nevertheless, these were added to the casualty figures, which compounded the level of security threat as Uyghurs continued to be publicly framed through discourses of danger and a threat to Chinese identity-security.

99 Based on interviews and conversations in the city at the time.

154

As abhorrent as the syringe attacks may have been, they represent attacks of the stigmatised and (re)performances of this stigmatisation. Xinjiang suffers from some of the highest rates of Heroin use and AIDS in the PRC (CSIS, 2007, p.3-4). AIDS and drug use are frequently stereotyped as a Uyghur problem. There have been periodic health scares revolving around these stereotypes in the region. In late 2007, the Ürümchi City Health Department sent a public SMS message stating that Uyghurs had strategically infected kebabs with HIV across the region100. On November 2011, rumours spread that 20,000 HIV-infected terrorists had laced food across China with their blood. These rumours were eventually publicly denied by the Ministry of Public Security and Ministry of Health in Beijing (People’s Daily, 2011). Uyghurs are being represented as sources of infection and threat to the health and security of the Chinese geo-body through unofficial and official security discourses. This medicalisation of ethnicity from above and below represents the fear that the cultural integrity of the unchanging core of the nation (Han-ness) is threatened with pollution through inter-ethnic exchange with Uyghurs. These representations shape how Uyghurs, who are marginalised by these discourses, resist these epistemologically violent discourses in physically violent ways. This is often inverted by Uyghurs who frame Han as the source of infection of the Uyghur nation. For example, Rebiya Qadir’s autobiography, Dragon Fighter, claims that the programme of resettlement of Han in Xinjiang was “an insidious one where all Chinese nationals who had become infected with AIDS were sent to our land” (Kadeer, 2009, p.273). However, these syringe attacks represented a different Uyghur discourse on identity politics. They were less an inversion of the text to articulate security and more of an embrace of the difference and violence which are officially attributed to Uyghurs.

The social dynamic which attributed group-ness to criminal acts and essentialised Uyghur-ness produces the stigmatisation of Uyghurs as a problematic group in Chinese nation-building. Ross Anthony showed how the daily lives of Uyghurs are subject to “heightened surveillance” by the security apparatus in Ürümchi in ways in which Han residents are not (Anthony, 2011, p.51-52). In the shopping area of Erdaoqiao/Döngköwrük, I observed armoured tanks patrolling the main square, half a dozen stationary trucks full of PLA troops, and troops at all of the main entrances with machine guns pointing over their sandbag defences. The positioning of police and military at this time exemplified the (in)securitisation of the daily lives of Uyghurs. Newly installed “electric eyes” in all underpasses, to the cost of 1 million RMB, confirmed the sense of “heightened surveillance” across the city (Chen Bao, 2009a). Close circuit television surveillance coverage across the Erdaoqiao/Döngköwrük confirmed its ethnicisation. Cameras were observed flashing on a near constant basis up and down the main roads of this area. This was

100 I was living in the city at the time and received this particular security SMS text along with all other mobile phone users in the city.

155 still observable after re-visiting the region in July 2012 and serves to reinforce the idea that Han areas are safe and Uyghur areas are dangerous.

The Han-centric life of the nation in Xinjiang was being articulated and securitised in the observable politics of the everyday. Every primary and middle school was required to have security guards and conduct security training for every student before the 60th anniversary of the PRC on the 1st October (Chen Bao, 2009j). According to multiple eyewitness accounts, many Uyghurs had been killed in the Qingnian Lu area and the Normal University in the north on the 7th July. Nevertheless, these areas were only lightly patrolled by neighbourhood watch patrols. These patrols were composed of volunteers, approved by their local work units (danwei) and residential committees101. From July to October, visible displays of power and violence were more widely observable in the south of the city, populated by Uyghurs. For example, the entrances to the 43rd and 32nd primary schools on Shengli Lu and Xinhua Lu were each cordoned off by paramilitary police. ‘China’ is being produced via the identification of insiders (Uyghurs) as a source of contamination to the life of the nation through their association with the outside (terrorism).

In the lead up to National Day, these schools were guarded for at least a week by PAP holding machine guns at the ready position. All bus stops, street corners, and mosques were guarded by 3-4 armed PLA grouped together. PLA patrols of about a dozen troops with machine guns, batons, and shields marched up and down every surrounding street. Uyghurs and their neighbourhoods were being designated as a danger. However, Han majority areas were being designated as safe and ‘inside’ with only a very light patrol of neighbourhood volunteers. The dichotomised approach to security not only reflected a perceived threat presented by Uyghur- populated areas but (re)produced it in ethnicised security practices, which gave visible markers of danger and security attached to different residential areas which all residents could observe. The security presence became a normalised part of the city’s background. Armed guards and security checks on all public buses became a publicly unchallenged and unquestioned part of daily life. This was comparable to the “unconscious hanging of the flag” in Billig’s idea of banal nationalism as a set of unconscious and taken for granted commitments to nationalism (Billig, 1995, p.6). However, ethnic boundaries defined through a hierarchy of inclusion were being taken for granted, not an all inclusive national model of equal citizens. It was taken for granted that Uyghurs need to be monitored for signs of exception because the criminal acts of particular individuals had been ethnicised and securitised beyond politics.

101 This is based on ongoing observations from September 2009 through to July 2010, where I would travel from the north to the south at least 4 times a week. Posters advertised for volunteers who, if selected, wore “patrol” armbands.

156

The performance of threats was accompanied by the mutually constitutive performance of the referents of security. Images of PAP officers armed with machine guns appeared in local newspapers alongside children raising the flag under the headline “Ceremony held, the People’s Police Protect the National Flag” (Chen Bao, 2009k). The nation as an idea and a form of symbolism was under threat and the threat of violence by the party-state was present to protect children in their ceremonial production of China. The city had been heavily guarded since the violence of July but this was scaled down by mid-September. North-South running roads were re-opened. Thousands of PLA troops relocated from the streets to make-shift barracks in public offices, including the Museum and the Regional Exhibition Centre. However, security was scaled up again in Uygur-populated areas for the celebration of national day on October 1st. This produced a Uyghur threat not only to security of the physical bodies of residents but to the idea and meaning of the nation itself. On the 1st October, a dozen PLA trucks containing approximately 30 troops, armed with machine guns, batons, and riot shields, paraded in front of the back-gate of Xinjiang University on Yan’an Lu. These were parked about 20 feet apart adorning slogans such as “defend the frontier” and “protect minzu tuanjie”. Uyghur areas were militarised to “defend the frontier” from the threat of Uyghur ethnic identity, as opposed to the security offered by the ethnic unity of Zhonghua Minzu. This designation of Han-ness as security and Uyghur-ness as a danger performed the internal boundaries of the nation in a hierarchical way so that to secure the nation it would need to become more Han.

The discourse of identity-security which performed victims of assault as national bodies created a larger sense of insecurity amongst the population at large as they felt their bodies and their nation was under threat at the same time. This resulted in more serious cases of ethnically targeted violence in response but which went unreported in Chinese media. According to reports in Radio Free Asia and by Uyghur bloggers, the singer, Mirzat Alim, was beaten to death by a mob in his home on the 2nd September and Kaynam Jappar, a well-known calligrapher and journalist, was being severely beaten by 6 Han Chinese men on the 3rd (Radio Free Asia, 2009b)102. In some of the cases reported in the local newspapers victims of syringe stabbings were able to chase and apprehend their attackers but what they did not report was what happened after they were apprehended. In one such reported case on the 3rd of September at Nanhu Lu, several non-Uyghur eye witnesses described to me three Uyghurs being dragged from a public bus and battered to death by a large mob of Han Chinese residents. They had been accused of a syringe attack and the mob prevented both the PLA and an ambulance from entering the scene until after they were both left for dead. The men may or not have been guilty but it is unlikely we will ever know as these murders were left unmentioned in the public narrative. This narrative performed the internal/external boundaries of the nation by framing

102 There is no information available on why these particular individuals were assaulted. It may have been because they were well-known. However, given that violence was ethnically targeted during July and September 2009, it is just as likely they were merely in the wrong place at the wrong time.

157 violence against Han Chinese bodies as a threat to national integrity. At the same time violence against Uyghur bodies was silenced and excluded from the narrative of the national geo-body.

4.3.2: Han Nationalism and Popular Protest

Securitisation heightened insecurity and its ethnocentric framings manifested in a popular nationalist rally on the 3rd September, which the party-state struggled to control. A huge protest of tens of thousands of residents, almost entirely consisting of Han Chinese residents, assembled across the city, with particular concentrations at the North gate (Beimen), the South gate (Nanmen), and the People’s Square 103 . The rally was described as a “protest against syringe attacks” in official state-media, which informed readers that protestors waved Chinese flags, shouting “severely punish the mob” (Xinhua, 2009L)104. Violence spread across the city on these days, which targeted Uyghurs in the social hysteria for revenge over the threat of syringe attacks. However, the official narrative did not relate the Han protests and this violence. The protests were framed as part of the discourse for security aligned with the party-state against the “inside/outside Three Evils”. This was in stark contrast with the representations of protests by Uyghurs with Chinese flags on July 5th for “justice” and “equality” which were framed as security threats.

The protests had emerged largely in response to the syringe assaults, which more than a dozen Han interviewees who participated and witnessed the protests confirmed. The media coverage which framed the assaults as threats to the nation shaped the panic and these protests. However, the official narrative also glossed over the multi-facetted nature of the protests and the competing understandings of security and the nation, which took place in a complex interplay between official and unofficial Chinese nationalisms. Chinese media did not mention the PAP’s use of tear gas to disperse the Han Chinese crowd 105. This conflict between Han residents and the security apparatus suggests a more complex relationship between state and society than the narrative of a China unified in struggle against ethnic separatists. International media and scholars reported the calls from the protest to sack the regional party chief, Wang Lequan (Millward, 2009; Moore, 2009). There were also chants to “kill Wang Lequan” (Wang

103 This was verified by three foreign eye-witnesses, several Uyghurs, and with at least 10 Han Chinese residents. These accounts varied in interpretation but offered near identical accounts of the empirical detail.

104 The international media repeated these framings accepted from the party-state, such as the Daily Telegraph’s Malcolm Moore in Shanghai stating this was about “complaining of lack of safety” after syringe attacks (Moore, 2009). However, the party-state control of the media ought to be considered here before drawing any conclusions. The vast numbers involved were not mentioned in local media such as the Chen Bao and reporting of the issue was seriously restricted. For example, three reporters had their cameras confiscated by paramilitary forces, others had tapes confiscated, and there were at least two separate instances of serious physical violence against reporters (FCCC, 2009f; 2009g; 2009h; 2009i). 105 This was mentioned in the FCCC reports and confirmed by several eye-witnesses

158

Lequan Shashasha) such was the level of anger amongst Han Chinese residents. 106 The international media and indeed many protestors interviewed during fieldwork framed this as an attack on the party-state’s lack of response to the syringe attacks. One Han Chinese interviewee in the Daily Telegraph was quoted as saying “the government is useless” (Moore, 2009). The New York Times quoted a retired Han Chinese woman, saying “The government hasn’t done anything, they haven’t told us anything. They haven’t kept order. We’re all so angry” (Wong, 2009). The PAP poster display of “7-5”, which had identified Rebiya as the “life or death enemy”, saw more vandalism to Wang Lequan than any other poster when Han Chinese protestors scratched and drew over his face (see figure 4.4). This suggests divisions and tensions between the party-state and the Han population over competing conceptualisations of Chinese identity- security. The hundreds of disappearances and arbitrary detentions discussed in chapter 4 were not sufficient to calm nationalists who were seeking to “restore the Han” through violent ethnic revenge for “7-5”.

The official narrative did not mention the need for thousands of PLA troops and PAP to block all major routes from the north to the Uyghur-populated districts in the South on the 3rd September. These stayed in full force for weeks after the incident to control the flows of people from Han-populated areas to Uyghur areas and vice versa. Groups of protestors attempted but failed to break past military cordons to reach the Uyghur district, Döngköwrük / Erdaoqiao. Several eye-witnesses, including several Han Chinese participants, explained how some protestors shouted “kill the Uyghurs” (weizuren, shashasha) alongside chants of “kill Wang Lequan”. The participants I interviewed were all self-employed or working class107. They raised their fist to me shouting “kill Wang Lequan, kill the Uyghurs” and expressed anger that they were stopped from reaching Döngköwrük / Erdaoqiao to “kill the Uyghurs”. Official Chinese nationalism with “the inside/outside Three Evils” as its significant Other was reinforcing the sense of superiority in Han nationalism but also (in)securitising this superiority. The Chinese security apparatus eventually had to prevent Han nationalists from taking security in to their own hands to attack the Uyghur enemy. This created feelings of injustice and marginalisation by these Han nationalists who as the “nucleus” of the nation felt they were securing the nation by maintaining their superiority over Uyghurs.

106 This is based on multiple eye-witness accounts and interviews with many Han who proudly discussed the protest. 107 This point will be covered in more detail in our final chapter, which utilises more detailed interview materials.

159

Figure 4.4: “Regional Party Secretary Wang Lequan makes his important speech regarding the ‘7-5’ incident”.

The protests were more complex than appeared and without survey data, we only have observations of events and scattered interviews to assess and interpret their meanings. It was clear that different people were involved for different reasons. One male, self-employed, middle-aged Han Chinese resident explained that the reason people took to the streets is that “the Han are angry”. He explained they are angry with unemployment, “inequality between Xinjiang with the rest of China”, and the “lack of security”, particularly due to the syringe attacks. Like Uyghur protests, this “snowballed” due to a broad range of social problems, for which Han Chinese blamed “the Government” and “Uyghurs”. Most participants interviewed expressed disappointment at the lack of executions of “Uyghurs” following the incidents of July 5th. More than a dozen interviewees cited security issues but all framed the protest as a Han protest using vocabulary such as “we Han” and “our minzu” to describe the participants. These framings indicated how popular perception had been shaped by the ethnicised and (in)securitising discourse of “The Three Evils”. “The Three Evils” discourse did not invent discrimination against Uyghurs. Nations are not imagined on a tabula rasa but are discourses built upon older discourses. However, the exclusionary discourse of the party-state exacerbates discrimination by representing Uyghurs as a threat to the nation. This discourse was then inverted against the government by protestors seeking to “restore the Han” and remove politicians from office. The grammar of nationalist legitimation was being mobilised as a

160 resource of Han Chinese resistance against the multi-ethnic model of ‘China’ offered by the party-state.

On the 3rd September the city government and the Public Security Bureau responded to the resulting disorder by pasting posters stating protests and rallies were illegal and that “traffic controls” were in place from 9pm 108 . The posters asked participants to allow “unity to be maintained by the party and the government”. This was a continuation of the Han-centred nationalist discourse that Han protests were for the unity of the nation despite the violence against and exclusion of Uyghurs sought by many of the protestors. By the 5th, Meng Jianzhu, Minister of Public Security, visited Ürümchi, stating that the goals of the protest, to “severely punish criminals”, had been met (Chen Bao, 2009p). This was an attempt to co-opt resistance to minzu tuanjie by Han Chinese and claim that their goals had been met by political leaders as guardians of the nation. This attempted co-option may have been politically efficacious but it also silenced the ethnic Han self-identifications which were being articulated counter to the party-state’s model of nation-building. This also failed to acknowledge, let alone resolve, the security threat that Han nationalists presented to Uyghurs who were becoming victims of physical violence.

Ethnic boundaries were being performed across the city. Different meanings were being attributed to different violences, which produced an ethnic hierarchy where Han-ness was being positioned above other minzu and represented as a trans-ethnic form of nationhood. The UHRP (2011) cite a video which shows PAP distributing large sticks to Han Chinese residents during the violence of July. By the end of September, I had observed dozens of neighbourhoods, particularly in the north of the city, with Han Chinese men aged from 18 to their 60s with large sticks and clubs. Official patrol volunteers were nominated by every residential unit but they were observably being joined by their “comrades” who paraded in the name of security but perpetuated the street-level politics of city where the threat of violence was ever-present. On Qingnian Lu (Youth Street), extendable batons, prohibited under European law, were on sale for 20 kuai at the front of an underwear shop. These were sold out in a day. I even observed several children playing on busy streets with these deadly weapons. Violence had become part of the banal and normalised background of city life.

As insecurity grew, martial law was restored. Tens of thousands of troops were posted to block main thoroughfares, most notably those running south towards Uyghur neighbourhoods. Jiefang Lu (Liberation Street) runs from the symbolic centre of the city, Nanmen (the old South Gate), all the way through the Uyghur district, Erdaoqiao / Döngköwrük. The whole route was

108 During this period many public announcements were posted across the city by official state bodies and they remained in place and easily observed for months.

161 fenced off and guarded with tanks, armoured cars, and approximately 1,000 troops armed with machine guns and riot shields. The troops faced north awaiting Han Chinese rioters. This could be easily observed but all mention was absent from the official narrative. This military division of the city into north and south took place exactly where the old Qing dynasty era city wall ran across what is now Renmin Lu (People’s Street). This boundary separated the small walled Manchu and Han cities (gongningcheng and dihua) connected in the late 1880s109 from the Uyghur and Hui Muslim city in today’s Erdaoqiao / Döngköwrük.

Ürümchi had only been seasonally inhabited by nomadic peoples until the 18th century with the arrival of the Manchu. These soldiers burned down the nearest indigenous settlement 5km away at Jiujawan and recognised its strategic importance in securing the “Western Regions” (Gaubatz, 1996, p.45-46). During periods of territorial expansion, Chinese people preferred to establish “new towns” based on the architectural ideals in the Chinese classics rather than settle in existing cities (Gaubatz, 1996, p.2-3). The “Frontier city” was built as an “outpost of Chinese settlement in non-Chinese territory” (Gaubatz, 1996, p.45). These “frontier cities” such as Ürümchi, , and Xining began as small Chinese enclaves but became multi-cultural cities as non-Chinese groups came to live in or near them (Gaubatz, 1996, p.2). These were military outposts of Chinese “civilisation” on the “new frontier”. After the Qing “unification” of Xinjiang, a policy of Han-Uyghur segregation was implemented. As more and more Han traders arrived in Xinjiang the new cities they inhabited became known as “Han cities” (Hancheng) whereas older cities became known as “Muslim cities” (Huicheng) (Pan, 1996b, p.86). The loose ethnic segregation in terms of residence between Chinese (Han) and Muslims (Uyghurs) which exists in Ürümchi today follows these architectural and civilisational fault-lines. The expansion of the Qing’s physical boundaries of China has never been fully resolved in the realm of self- identification: “just as to Han minds, Xinjiang was in the empire but not yet of it” (Millward, 1994, p.442). The militarised and ethnicised performances of this ancient civilisational boundary between Turkestan and China reproduce Xinjiang’s position as an unresolved frontier. The state has been made but a national community is not yet built. The boundaries between Han and Uyghurs continue to be organised violently and hierarchically. Xinjiang remains a “new frontier” today.

At the end of April 2010, Wang Lequan was removed as party secretary for Xinjiang. This was a post he held for 15 years and in which he became popularly known as the “King of Xinjiang”. Wang was a sacrificial pawn to appease those protestors who sought to expel Uyghurs as a “race of demons”110. However, he was moved to the position of Deputy Chair of the Political and Legislative Affairs Committee in Beijing, where he is head of the influential Xinjiang Work

109 I am indebted to the archival research of Eric Schluessel for pointing out that the Han and Manchu cities were connected at this point in history. 110 The final two chapters will analyse popular discourses on ethnicity and danger using interview materials.

162

Team. This move was never fully explained by the party-state as is often the case with political reshuffles in China. It was clearly responsive to social demands from Han protestors. However, it is doubtful that it will have any substantial impact on policy if Wang remains in an influential position and the regional government remains subservient to the centre. The party-state continues to produce the nation in Xinjiang through the double-edged sword of Chinese nationalism. Official nationalism does not cause popular nationalism but it does inform it by providing a vocabulary and ontology which can be (re)performed for competing ends. Official articulations of identity-security shape how citizens attribute particular acts of violence with particular meanings and securitise their identities in ways the party-state cannot control but to which it has to respond. On the anniversary of July 5th in 2010 the streets and public transport were largely empty of residents for most of the day. This indicated the fear which remained in the city. By July 5th 2012, the city was considerably more bustling. However, the pattern of armed PAP patrols, the constant flashing of surveillance cameras, bag-checks, and temporary security gates in Uyghur areas all remained in place. The ethnicisation of security and the securitisation of ethnicity appears as if it will continue to shape the nature of inclusion/exclusion for some time. The popular streams of Han nationalism have been shaped by the party-state’s discourses on Xinjiang. However, the flow of influence is not only one way as they then shape the party-state’s model of a multi-ethnic China with a Han-centred discourse of minzu tuanjie.

Conclusions

Xinjiang is represented by the party-state through exclusion from the centre as a “frontier” on the periphery of “China”. However, inclusion is central to the ongoing production of Zhonghua Minzu as a bounded state of 56 unified ethnic groups who self-identify as Chinese. This chapter showed how Chinese nation-building in Xinjiang articulates both the internal cultural boundaries of China (domestic) and the external boundaries of the state (international). Inclusion and exclusion are theoretically indivisible aspects of boundaries and their ongoing production. In Xinjiang, this takes place through the identification of cultural enemies within China and linking them to a dangerous ‘outside’ of “international terrorism” and “Western” “capitalism”.

Section one re-orientated the concept of securitisation through Judith Butler and Cynthia Weber’s concept of performativity. It applied this theoretical framing of identity politics to the official explanations of the violence of July 2009. Through the discourse of the “inside/outside Three Evils” these explanations were performative acts of boundary production. However, this unintentionally produced antagonisms between different groups and created further insecurity between them. We saw how resistance to the party-state was not external to the power of the party-state. Protestors drew on key Chinese symbols, the national flag and “equality” to

163 performatively enact their desired position in China. We saw too how Han nationalism was not controlled by the party-state but that they both shape each other, producing and responding to the social antagonisms with Uyghurs, which lead to mass violence.

The second section showed different meanings were attributed to violence by different groups such that violence by Uyghurs was securitised as a threat to China where violence by Han was either ignored or represented as an ordinary issue of “public order”. These meanings produced an ethnicised dichotomy of subject positions where Uyghur self-identification through language and religion were framed as dangers from the ‘outside’. Han-ness was conversely represented through Chinese security and inside-ness. These discourses attached different ethnicised values to different bodies. The different meanings attributed to violence shaped how different groups were treated by the security apparatus, particularly through the practice of arbitrary detentions. It was then shown how China’s ‘domestic’ policing based on its model of identity-security was applied to international politics. This produces a marginalised position of Uyghurs at home and in the world. Uyghurs who flee China are subsequently deported and imprisoned without proper trial. The final section outlined the aftermath of the violence of July. It analysed how the dichotomies of security/danger and ethnic/national continued to be performed in official discourse and (re)performed in popular Han responses to the violence. Han protests were against the government but for the Han nation in a move which complicates the relationship between Han and the state. Han in Ürümchi were rejecting the party-state’s model of a multi- ethnic nation by ethnically target Uyghurs. However, they are doing this in ways which draws on but inverts the discourse of the “inside/outside Three Evils” as an ever-present threat to the nation.

This chapter showed how security and identity are inseparable discursive practices whereby “who are we?” and “what is to be secured?” are intimately linked questions. The answers to these questions emerge through social practices at the level of the state in interaction with social groups. Nations and ethnic groups like all social categories are discourses within which competition and contestation over their meaning and boundaries takes place. This contestation is a dynamic process of ongoing production of these categories. This contingency and contestation which the category conceals is not a problem in itself and it can involve members of groups as participants in the making of identity politics. However, in our case, the Chinese party-state faces an internal/external security dilemma. The party-state demarcates boundaries which exclude how postulated members of the nation identify themselves and then includes them in hierarchical ways. The party-state represents the security dilemma it faces as an internal problem and one which should not be internationalised. However, its approach to identity politics frames Uyghur-ness as Uyghurs define it through “discourses of danger” linked to international “enemies”. This superimposes cultural boundaries within China upon external

164 boundaries whereby Uyghur-ness is represented through discourses of “international terrorism”, “western capitalism” and the interfering international media. This has exacerbated security problems in the region where Uyghurs resist the re-organisation of their inside and outside and frame China as the threat to their identity-security from the outside. The party-state securitises a model of Chinese-ness which one group (Uyghurs) understands as assimilation into another group (the Han). However, the other group (the Han) frame this as benevolent incorporation. This further exacerbates tensions between groups. This decreases security in cycles of violence and insecurity which contest identity categories and in which the state as the arbiter creates further insecurity by targeting one group. The next chapter will show the inclusionary discourse of minzu tuanjie in the politics of the everyday objectivises ethnic Han-ness. Han-ness is equated with Chinese-ness, which contributes to further feelings of exclusion, assimilation, and insecurity amongst Uyghurs.

165

Chapter 5: Performing Uyghur Inclusion, Objectivising Han-ness

Introduction

Chapters 2 and 3 examined the exclusionary representations of Xinjiang as a dangerous “frontier” through the discourses of East Turkestan and the “inside/outside Three Evils”. Nation- building seeks to displace the existing Turkic and Islamic self-identifications of Uyghurs, which are marked by the party as dangers to be overcome. This chapter shifts focus to the inclusionary dimensions of nation-building as the mirror-image of these models of exclusion. It will outline what type of nation China is performing in Xinjiang by exploring the ways in which Uyghurs are included as Chinese. The analysis will focus on how the inclusion of Uyghurs in China was performed by the party-state through representations of key political and cultural events in Ürümchi in the year following the violence of July 2009. These performances are the promotion of Minzu Tuanjie education and the public celebration of traditional ethnic festivals of the Han (Zhongqiu jie) and Uyghurs (Roza Heyti). The aim is to uncover the model of inclusion and articulation of a national self that securitisation offers by excluding particular characteristics as threats and unbefitting of the community.

The first section will focus on the inclusion of Uyghurs in Chinese nation-building through the public performances of minzu tuanjie and Zhonghua Minzu in Ürümchi during 2009-2010. This will explore how inclusion of Uyghurs China is performed in the politics of the everyday through publicly played songs and personal displays of national emblems. Inclusion through minzu tuanjie is hierarchical because Uyghurs are not being included in modernity or the “nucleus” of the nation. Xinjiang is included as a frontier, which needs to be developed, and Uyghurs are included as peripheral to the Han “nucleus”. The second section will analyse the representation and celebration of different traditional ethnic festivals in the city in 2009 to 2010. This section will analyse how these festivals were represented through the prism of the ethnic/national dichotomy in ways which normatively order identities into a social hierarchy. These celebrations offered a model of hierarchical inclusion by framing specific festivals as ethnic and others as national and trans-ethnic. Through the example of minzu tuanjie, this chapter will show how nation-building in Xinjiang objectivises ethnic Han-ness as national Chinese-ness thus offering a model of conversion of difference rather than its inclusion. This model of inclusion and the ongoing role of the state and the military in the economy and society of Xinjiang are significant to how we think more generally about China. The nature of inclusion in Xinjiang challenges the conventional narrative that China, as a singular unit of analysis, is on a linear path towards

166 modernity and open-ness. Instead, Xinjiang shows that open-ness is highly uneven across China and thus that more regional and local analyses can help us better understand ‘China’.

5.1: Hierarchical Inclusion/Exclusion and the Struggle for Minzu Tuanjie

This section turns to the inclusion inherent in nation-building through public performances of Minzu Tuanjie and Zhonghua Minzu in Ürümchi. It will examine how the performative inclusion of Uyghurs through minzu tuanjie and the exclusion of “The Three Evils” in the politics of the everyday are inter-textually connected in a process of mutual constitution. The analysis will explore how inclusion and exclusion should be understood as inseparable aspects of building a socially hierarchical nation. Minzu tuanjie and “equality” have been described by the party as the “core content” and “cornerstone” of China’s ethnic minority policies in the reform era (State Council, 2003, p.79; State Council, 2009a, p.14). Following the violence of July 2009, the party- state and the regional government promoted a region-wide drive for compulsory and universal minzu tuanjie education. These materials are used to “promote ethnic unity consciousness…with patriotism as its core content” (XUAR Government, 2009, article 3). Official documents and the education materials explain how this policy is located within the broader drive for “patriotic education” and nation-building111. Minzu Tuanjie is a theory premised on the timeless unity and territorial enclosure of ethnic groups. Throughout history, many of these groups have been militarily and politically antagonistic towards each other. However, today they are re-represented to desire their own destruction through the formation of a Chinese national identity (Bulag, 2002, p.1 & 22).

Minzu tuanjie is the “fundamental principle” in dealing with the “minzu question” (minzu wenti) such that ethnic identity is to be superseded by a Chinese national self-understanding (State Council, 2009a, p.14). The universal coverage of this policy enables the broad dissemination of the party-state’s model of nationalism. Just as debates on minzu policy in chapter 2 focused on national strength, the party-state stresses that all multicultural countries need minzu tuanjie if they are to be “unified”, “stable”, and “developed” (State Council, 2009a, p.1). Hu Jintao declared it as central to national stability and the “great revival of the Chinese people” and as we saw in earlier chapters it was linked to prosperity of the whole nation defined against the “disaster” of separatism (Hu, 2010). Minzu Tuanjie then is taught to students as the basis of the “great revival of the Chinese people”, the basis of “national strength”, and the expansion of China’s international “soft power” (Ministry of Information, 2009b; Ethnic Unity Education Board, 2009). Minzu tuanjie illustrates the inseparability of internal and external boundaries whereby domestic identity politics are embedded in discourses of international prestige and power. Minzu tuanjie aims to produce ethnic extinction, which is drawn by Ma Rong (2007; 2009) from his

111 For example, see: XUAR Government (2009); Ethnic Unity Education Board (2009).

167 interpretations of US identity politics. However, this model of inclusion demarcates an international boundary such that China’s historical experiences can only be understood through domestic, multi-ethnic tuanjie while the “West” has its separate experience of nation-states. This is a way to show that China attracts rather than assimilates ethnic groups, thus deflecting criticisms from policy toward ethnic minorities. However, it also performs Chinese-ness by showing how people must both identify with the attraction of the Han centre (inside) and not with the “West” (outside).

James Leibold theorised three modes of ethnic inclusion in China: state-sponsored Leninist minzu policies, Han , and Confucian ecumenism (Leibold, 2010, p.2). Leninist minzu discourses were most apparent in the analysis of ethnic classification and “scientific development” as means to “modernise” minorities in chapters 2 and 3. Confucian ecumenism, which entails cultural fusion of all minzu into ‘China’ was explored in the works of Chinese scholars, particularly Fei Xiaotong and Ma Rong. Civic and cultural nationalisms are not mutually exclusive and even in paradigmatic cases such as France and the USA ideas about culture are often constitutive of statehood (Brubaker, 1999, p.60-61). The analytical separation of the civic from the cultural does not help us understand the power relations which bestow the capacity to label particular social practices as civic. Those nationalists, who represent nations as civic, thus non-cultural, are participating in social practices which conceal assumptions of the normality of the majority culture and the anomaly of the minority.

The idea of a civic, multi-ethnic Chinese state conceals the essentialising ethnic ontology behind this discourse. Fei Xiaotong’s (1988) idea of the “Han nucleus” and Ma Rong’s (2007) idea that minorities should be culturalised (i.e. a non-civic identity) are leading examples of this ontology in China. Cultural characteristics such as Confucianism are thought of as products of an imagined biological group today known as hanzu (Dikötter, 1994, p.404). Rather than see Leibold’s three models as mutually exclusive categories, the argument here is that all three are mutually reinforcing strands of nation-building discourses in Xinjiang. These are all forms of Han-Chinese nationalism. In each discourse the Han is positioned as the centripetal centre of the nation as a source of modernity (Leninism), cultural attraction (Confucianism), or cultural superiority (ethnocentrism) within the confines of the territory of contemporary China stretched back to “ancient times”. The drive to categorise minzu (Leninism), unite them as Chinese (Confucianism) but place the Han at the centre (ethnocentrism) are of equal prominence in Chinese nation-building. Communism then did not supplant Confucianism but became transplanted onto existing ideas of Chinese-ness where the “nucleus” became the “vanguard” and in the reform era it has become a source of “development”.

168

In Xinjiang, news broadcasts, school textbooks, and political slogans analysed in previous chapters are saturated with the discourse of minzu tuanjie in such a way that it is “everywhere” in the sense Foucault (1991) intended in Will to Knowledge. Understanding minzu tuanjie thus requires we look beyond the ‘high’ politics of party conferences and explore the texts which people encounter in the ‘low’ politics of the everyday. One textbook used in minzu tuanjie classes explains that one of the key precepts of minzu tuanjie is Jiang Zemin’s concept of “The Three Cannot Leaves” (sange libukai). The concept of sange libukai emerged prior to the Cultural Revolution. Following its disuse, Jiang Zemin re-introduced the idea during his 1990 visit to Xinjiang. It now appears in all party documents and educational materials on ethnic relations: “The Han can never leave Shaoshu Minzu, Shaoshu Minzu can never leave the Han, and all Shaoshu Minzu can never leave each other” (see figure 5.1). In this case, the advert for “boundary-crossing” MG Cars appeared alongside the “three cannot leaves” in a urinal in an Ürümchi coffee shop. The Han and minorities are imagined as one indivisible national geo-body. The everywhere-ness of this discourse can be seen where identity politics and commercial appear not only in competition for public space but alongside and merged into another.

Figure 5.1: “The Han can never leave Shaoshu Minzu, Shaoshu Minzu can never leave the Han, and all Shaoshu Minzu can never leave each other”.

169

Minzu Tuanjie emerges from relations of power between the state and different groups, which define who people are. What counts as a minzu and a nation reflects the power relations which define the distinction and relationship between ethnicity and nation (Pan, 2008b, p.112). However, these power relations also shape the meanings attributed to different minzu, in our case Han and Uyghur, and not just the abstract category of minzu itself. This form of inclusion based on what Gayatri Spivak (1988) called “epistemic violence”. Minzu tuanjie speaks for and defines the boundaries of different groups without allowing them to speak. This inclusionary Othering was explained by Louisa Schein (2000) as “internal orientalism” where ethnic minorities are marginalised as “backward” but framed as inseparable components of the Chinese nation at the same time. This model of inclusion imagines different positions in the social binarism to correspond with different positions of cultural evolution discussed in chapter 2. Drawing on Fei Xiaotong, the notion of the Han as the centre of the construction of the Chinese nation appears alongside sange libukai in the party-state’s literature. Zhonghua Minzu is then primarily based on the timeless, unbroken culture of the central plains as centre and Xinjiang as a peripheral “frontier”:

“Sange libukai reflects the history of the progress of minzu tuanjie amongst all of China’s minzu. Over thousands of years of history the Han of the central plains and the surrounding area’s shaoshu minzu have assembled, becoming the unified and stable Chinese nation with the central plain as the core…the Han are superior in the areas of their economic cultural level, science and technology, and their labour resources” (Ministry of Information, 2009a, p.43-44).

Xinjiang is represented as a “meeting point between savagery and civilisation”, or as argued here, between China (hua) and the outside (yi). However, unlike other ethnic minority regions, Xinjiang is marked as a security problem through the discourse of the “inside/outside Three Evils” whereby “savagery” is not simply represented as “backward” and in need of modernisation but as a threat which requires exclusion and conversion. There is little worth in comparing levels of development to periods of ancient history. The more pressing analytical concern is how the party-state positions itself and the Chinese nation as the taken-for-granted cause of this development. This is an attempt to legitimise the CCP’s political control of the region as an essential source of stability and prosperity for the peoples of Xinjiang. This legitimation is an articulation of hierarchical identity politics and dependency. Minzu tuanjie is thus inherently hierarchical with the Han as the centre driving the economic and cultural development of China. This fails to address the identity-security concerns of Uyghurs that its goal is assimilation. Minzu tuanjie will more likely exacerbate these concerns because it includes Uyghurs in China as behind the Han on a spectrum of Chinese-ness defined through modernisation along which they are expected to advance.

170

The fear of savagery as a security problem and the desire for civilisation explains why the party-state’s response to the violence of July 2009 was to universalise minzu tuanjie education to convert people who identify as minzu into people who identify as nationals. By the end of 2009, the local newspapers in Ürümchi printed personal, “people-centred” “minzu tuanjie friendly stories” on a daily basis under the heading “My Minzu Brothers and Sisters”. These included the story of the “Uyghur Mother”. The “Uyghur Mother” was a housewife without children of her own, who helped a Han neighbour by looking after his children when he went to work (Chen Bao, 2009q). Another story was that of a young Hui man who arranged his for a date so that a Uyghur friend could attend (Chen Bao, 2009r). These stories were guided by introductory texts explaining how different ethnic groups are “one family” engaged in “common ethnic unity struggle and prosperous development”. The framing of these stories ethnicised and politicised everyday experiences into politicised acts of nationalism. Such representations were visually displayed in the free exhibition in the regional expo centre in Ürümchi, September 2009. This was organised and curated by the Xinjiang Uyghur Autonomous Regional Government for the 60th anniversary of the establishment of the People’s Republic of China.

Figure 5.2 appeared in the section “Tuanjie and Harmony Stories” (tuanjie hexie pian). This represented the “Uyghur Mother” as loving of all minzu and integrally bound into the Chinese geo-body. Peace and love were being ethnicised as much as the violence we analysed in our previous chapter. Like the victims of violence, the ethnicity of the heroes was highlighted. This ethnicisation allowed their actions to be framed through minzu tuanjie to performatively enact a unified and harmonious multi-ethnic nation. The framing of the actions through the prism of Chinese nationalism was an ideological imposition mobilised in the drive for nation-building. These representations dehumanised these actions because rather than becoming stories of humans saving other humans’ lives, these actions became framed as intrinsically ethnic and mobilised as evidence of minzu tuanjie. News stories and public exhibitions mobilised these images but did not give the participants a voice to speak for themselves in telling their own stories.

The mother-figure plays a huge role in Uyghur conceptualisations of community and in Uyghur popular culture. For example, the song “mother-tongue” (Ana Til) by Abdurehim Heyit was recreated in a disco-pop style song by Yasan Mukhpul. He sung “the language went into my body with the first drop of my mother’s milk…you are so precious to me like my dear mother”. Uyghur-ness is represented as transmitted via the all-powerful mother figure bestowed with both the responsibility of maintaining nature and the community. The importance of women’s reproductive functions in nationalist discourse is often tied to those which posit a “myth of common origin” in the construction of identity (Yuval-Davis, 2008, p.26). The mother-figure

171 then represents the natural, unbroken lineage of Uyghur-ness in modern China. This is what Duara (1998) called the “soul of tradition within modernity”. However, this imagery is used to resist the hegemonic orthodoxy of Chinese nationalism which seeks to articulate gender and ethnicity for Uyghurs. Unlike Uyghur discourses discussed in the previous chapter, the “Uyghur mother” deployed by the party-state is not the source of continuity of the Uyghur nation. She is the loving guardian of the multi-ethnic youth of Xinjiang and a guardian of minzu tuanjie.

In Xinjiang, a site of near-permanent “crisis” for the Chinese state, representations of women are embedded in nation-building discourses of conversion. These narratives integrate Xinjiang as part of China but as an Othered and economically “backward” region. Xinjiang, like other “shaoshu minzu regions”, is in many ways represented through women. For example, in early 2012, the Xinjiang oil-paintings exhibition, “The Affection for Motherland”, toured the Chinese cities of Beijing, Shanghai, Guangzhou, and Ürümchi to much fanfare on Xinhua. The project was recommended by , party secretary for Xinjiang, and organised in conjunction with the state-owned China Poly Group Corporation to “propagandise and eulogise Xinjiang” and “to show a beautiful, prosperous, harmonious, and confident Xinjiang” (Hu & Shao, 2012, p.3-6). This beauty and harmony was largely presented through representations of women. Of the 90 portraits on display split roughly between Xinjiang and “inner China” (neidi) painters112, 79 centrally or solely featured women or girls. Despite this being a showcase for “prosperous” Xinjiang, every artwork depicted women wearing traditional or “ethnic” clothing and only three of these artworks depicted Uyghur women in an urban setting. These representations and absence of urban representations articulate Xinjiang through the female and the rural, including it through “affection for motherland” but excluding it from the active and modern “Han nucleus” of China. These representations are mediated by the party-state from above but they emerge through market demand from below. In China’s new marketplace, Xinjiang is not represented as in a position of competition with other regions but through permanent marginalisation as a passive, rural, and feminine “frontier” to be developed. Following an incident of unrest in Yecheng in March 2012, Zhang Chunxian stated that “we shall show no mercy to these terrorists…we shall not let them wave knives at our women, our children, and our innocent people” (Xinhua, 2012a). Xinjiang, like its women are routinely represented through one another as objects to be secured and weaknesses to be protected.

112 I attended the Beijing exhibition and the book cited here features every painting from the exhibition.

172

Figure 5.2: “Alika Kedehan, ‘Loving Mother’ ”.

Another form of popular media which mobilised dozens of ethnicised images of these actions was the video for the pop ballad “One Family” (Yi Jiaren) produced by the Regional Government Party Commission Information Department and the Bureau for Culture (XUAR Party Commission Information Department, 2009b). Yi Jiaren was performed by host of singers from multiple ethnic groups113 and the writers claimed the music was based on a traditional Uyghur song. From September 2009 until summer 2010 the video was played every evening on Xinjiang television under an officially assigned “minzu tuanjie broadcast”. The song, like all messages of minzu tuanjie, was everywhere in the city for a period. It could be heard every day from loudspeakers on university campuses during break-times and the minzu tuanjie classes sung the song and for their examination had to write an essay explaining how the song made them “deeply feel about minzu tuanjie”114. The song was performed at the Xinjiang People’s Theatre and beamed back to (CCTV) for a national audience in Beijing for Spring Festival (Chunjie) during February 2010 (Xinjiang Dushibao, 2010, C03). Official explanations posted on Chinese video hosting sites explained the background to the song as the regional government’s drive to use arts to “promote tuanjie and protect stability” after “7-5”, which was instigated by the “inside/outside Three Evils” (Xin Kuan, 2009). The song lyrics (see

113 These were Han, Uygur, Kazakh, Mongolian, Hui, Kyrgyz, and Tajik. 114 Based on interviews with student participants.

173 appendix 1 and 2) propagated a representation of China as a unified, multi-ethnic family. This performed China not simply as a civic political unit but as an organic, intimate, and unbreakable familial bond. The chorus sang:

Home is one home, the country is great China, If the family lives in harmony, all affairs will prosper, It has you and it has me, Home is one home, the country is great China, We are all one family, you cannot separate you and me. Warmth expels the cold winds, true love becomes a true heart, We are all one family under the blue sky, We are all one heart, all one root, Prosperity accompanies peace, we are all one family115.

This was a rousing and emotive song, which sought to overcome the trouble of July (“cold winds”) with the “warmth” of the nationalist impulse of minzu tuanjie. This performed the nation (“great China”) through unified history (“one root”), identity (“one heart”), and development (“affairs will prosper”). The nation was performed as an organic and indivisible whole (“one family”), from which separation was less impermissible than it was logically impossible (“cannot separate”). The video showed the many performers passionately singing, interspersed with images of post-“7-5” unity. Particular prominence was given to images of donations of money to help ‘China’ return to prosperity. There were also donations of blood to injured residents in hospital, forming a literal organic bond between residents who were now sharing life from the “one heart”. The tuanjie between different minzu in the video then shifted to the unity between people and the state. Donations of food by the truckload were handed to military forces. The PAP in the video appeared armed with the machine guns they used in the divisive, ethnicised security practices across the city. Under the auspices of minzu tuanjie, they marched to the line “the country is great China”.

The final scene in the city performed the officially designated role for Uyghurs and other Shaoshu Minzu in China as the non-modern Other. Half a dozen ethnic minorities dressed in brightly coloured nylon “ethnic” costume walked against the urban backdrop of the city. The song articulated the relationship between different ethnic groups as unproblematic (“I don’t need to ask, you don’t need to say”). The video illustrated the relationship between Han as modern and Shaoshu Minzu as traditional. Despite the seemingly unproblematic nature of this relationship, the multi-ethnic nation emerged as an enforced peace through the threat of violence. The backdrop to the city at the time of the song’s wide exposure remained heavily

115 See appendix 2 for full lyrics.

174 militarised and the security services remained in position to police Uyghurs. The song was mocked by Uyghur students on university campuses. When there were no teachers around, they sang the song in faux high-pitched voices followed by roars of laughter. The nationalist message of unity was not being received by its target audience as intended. This was because the song ignored the daily practices of the securitisation of ethnicity, which ordinary residents could not avoid.

The inclusionary message of minzu tuanjie in the song Yi Jiaren was mutually constitutive of the exclusion discourse of the “inside/outside Three Evils”. It was produced by the department responsible for education materials such as The 50 Whys, which trained students and workers to defend themselves against “separatist thought”. The backdrop to the video was the aftermath of the violence of July. Students of The 50 Whys were urged to “unwaveringly struggle against the Three Evils, self-consciously protect national unity and minzu tuanjie (Ministry of Information, 2009, p.68). Exclusion and inclusion are theoretically indivisible and the struggle against “The Three Evils” is a struggle for minzu tuanjie:

Our struggle against the Three Evils is a struggle between protecting national unity against national separatism and of protecting minzu tuanjie against minzu separatism. It is a zero-sum political struggle of life or death (ni si wo huo). (Ministry of Information, 2009, p.15)

Minzu tuanjie links the power and prosperity of the nation through the “great revival of the Chinese people” (fuxing) and the struggle for development. The inclusion and peace it offers are mutually constitutive of the exclusion and violence reserved for the “inside/outside Three Evils”. This mutual constitution was evident in the images of state-violence and social harmony Yi Jia Ren video but also in the “Harmony Tuanjie Stories” section of the 60th anniversary exhibition. A photo of “the mighty army” grimacing with raised swords representing the threat of violence to enforce tuanjie (see figure 5.3) hung alongside an image of doves of peace flying across the People’s Square (see figure 5.4). Violence and peace did not so much contrast against each other as much as they made each other possible. Violence by the state to impose an ethnic hierarchy was being represented as a form of peace. Violence against the enemies of “The Three Evils” was intended to produce peace towards the friends of minzu tuanjie in a single interlinked “struggle”.

175

Figure 5.3 (left): “The mighty army – the troops of Xinjiang’s People’s Armed Police”.

Figure 5.4 (above): “Ürümchi People’s Square”.

Differences in the celebration of national unity between Han and Uyghurs were observed at this time. One such difference, which remained for months across the city, was the unequal adornment of an “I love China, I love Xinjiang” stickers and badges (see figure 5.5). The stickers were made by the Ürümchi Evening Times (Wulumuqi Wanshibao) to commemorate the 60th anniversary of the founding of the PRC. The newspaper explained that the anniversary had “raised the self-confidence of the Chinese family and exhibited the greatest opportunity of the new China’s great achievements” (Wulumuqi Wanshibao, 2009). The idea of these symbols was to “wish happiness on the great motherland and a more glorious future for our beautiful Xinjiang” (Wulumuqi Wanshibao, 2009). Furthermore, the distribution was claimed to be a response to the popular demand of residents. The newspaper explained that since the violence of July, “every member of the Chinese family” “wanted to express their warm love and blessings for the motherland”; “all children of every minzu will all feel pride and self-belief because they are Chinese” (Ürümchi Evening Times, 2009). The stickers thus demanded that love of China and of Xinjiang go hand in hand. Their embrace and rejection can then be seen as an indicator of the appeal of Chinese nationalism.

The Ürümchi Evening Times distributed 50,000 of such stickers and car badges at People’s Square, the North-gate (Beimen), Hongshan, Youhao Lu, Nanhu Square, and the railway station (Wulumuqi Wanshibao, 2009). Other than the railway station, none of these areas were in the Uyghur-populated south of the city. These stickers were being displayed on about half of all businesses in the city centre and the Qingnian Lu area. They were barely noticeable in the south. Over this whole period I only observed one Uyghur teacher at Xinjiang University

176 wearing the badge. I even had to explain what they were to several Uyghur informants who lived in the south of the city but had not even seen them. This banal nationalism and the boundaries which it elucidates were being performed by Han Chinese and not Uyghurs. Uneven participation in this imagined community reflected the lack of social connections between the north and the south of the city. Uyghurs did not want to be included in these performances of China. This uneven participation also reflected the differing community perceptions of national belonging and the ‘China’ concept.

Figure 5.5: Patriotic stickers given out across the city of Ürümchi: “Love China (Zhonghua), Love Xinjiang”.

The choice of Zhonghua instead of Zhongguo to represent ‘China’ on the badge was significant for two reasons. The term, Zhonghua, is seen by the likes of Ma Rong as the ancient term for China instead of its modern name, Zhongguo. Zhonghua then performs a timeless national community which runs from ancient times to today. Secondly, Zhonghua is an ancient term used to describe the Yellow River basin and the central plains as the place of origin of the Han116. This unconsciously objectivises Han-ness as Chinese-ness. It should be noted that the very name for the Chinese nation utilises this elision between Han and China 117 . This interpellates a Han-centred nation. Hua / Yi was the distinction between those on the inside of

116 See Ma (2007) and Fei (1988). 117 This includes the formal name for the PRC (Zhonghua Renmin Gongheguo) and the term used for the Chinese nation as a people (Zhonghua Minzu).

177

“the core” of Chinese civilisation (Hua) and the barbarians (Yi) on the periphery who “needed to be educated” (Ma, 2007, p.5). Rather than think of Hua / Yi as a fixed ontological reality, the approach here follows Cynthia Weber (1998) to analyse this language as a performative enactment of these socially contingent boundaries. These symbols were not a reflection of an existing identification with Zhonghua Minzu. They are a means to produce China as an internally hierarchical social order with the Han at the centre and Xinjiang on the periphery. Xinjiang and the Han of the Yellow River (centre and periphery) were being performed as mutually constitutive components of one another such that one could only love both for they are inseparable.

As “all children of every minzu” must wear these badges “because they are Chinese”, these badges performed Chinese-ness as a national practice for everyone to participate in but included at different levels of civilisation. The badges were then spread across the city and large companies took advantage of the patriotic sentiment by displaying the logo. For example the widely used television company, Xinjiang Guangdian Wangluo, displayed the logo on their cable television menu for several weeks. The social hierarchy of the Chinese nation was being publically performed in ways which became impossible to disentangle the official from the unofficial and the political from the everyday. When I asked young and old Han Chinese residents wearing the badges why they wore them, they had little to say but that they “love China, love Xinjiang”. Residents were actively participating in a powerful and unconsciously structured symbolism. This is best understood as a banal and taken for granted form of nationalism but one which concealed a complex and hierarchically structured understanding of what it meant to be Chinese. These social practices concealed an unconscious Han-centric vision of the nation in Xinjiang. Han-centrism was drawn from the popular Han nationalism in the region but holds little appeal for Uyghurs. The next section now turns to this objectivisation.

5.2: Ethnic Festivals and National Festivals: Marginalising Minzu, Objectivising Han-ness

The 1980s saw a period of “cultural liberalisation” in China where official conceptualisations of minority cultures were “fostered, promoted, and marketed” (Schein, 2000, p.88). ‘Ethnic’ festivals became reworked spectacles “clogged with tourists” (Schein, 2000, p.89-90). However, this section will show how different traditional festivals were represented and celebrated in Ürümchi from 2009 to 2010 in very different ways. A two-week period from the end of September to the start of October witnessed celebrations for National day, the Han Chinese Mid-Autumn festival (Zhongqiu jie), and the Uyghur Roza Heyti. The close proximity of these festivals in space and time provided an opportunity to analyse and observe how they were differentially framed and understood by Han and Uyghurs. This section will argue that Uyghur

178 festivals are marginalised as ethnic and relegated to the private sphere. However, traditional Han festivals are objectivised as national and trans-ethnic by being publicly promoted and celebrated. This challenges how we think about ethnic inclusion in China because in Xinjiang traditional festivals are being publicly silenced and removed from visibility more than they are being co-opted. This silencing exacerbates fears amongst Uyghurs of assimilation into a Han- dominated China.

The prism of the ethnic/national dichotomy offers a hierarchical model of inclusion. The discourse of Zhonghua Minzu and life in China is regularly characterised by Uyghurs as Han- centred or Han-dominated. This does not contradict Fei Xiaotong’s idea of a “Han nucleus” rather it (re)performs its meaning whereby this centre reflects domination and assimilation rather than harmony and attraction. This “unchanging core” is best understood as a “regime of authenticity” which imagines a unified actor progressing through time in ever-changing circumstances (Duara, 1998, p.291). The contingency of this unchanging authenticity is revealed by shifts in CCP policy. Policy has been transformed from the revolutionary politics of eliminating tradition to endorsing and celebrating traditional festivals in recent years. The CCP has diversified its sources of legitimation by appealing to the cultural nationalisms associated with traditional Chinese festivals over socialist holidays such as International Worker’s Day. The party-state is using festivals to contain resistance to its idea of ‘China’ by re-imagining itself and Chinese tradition as dependent on one another. The promotion of festivals at the national level allows the centre to contain rival administrations at the provincial and local levels of government. Provincial governments earlier sought to assert regional identities through the regional timing of holidays (Callahan, 2010, p.192). However, Uyghurs continue to articulate alternative regimes of authenticity by challenging the authenticity of the party-state’s regime. Examples of this resistance could be seen on Uyghur Online, an online message board and blog hosting site for Uyghurs using Mandarin Chinese. Uyghur Online was closed down as part of the “strike hard” against “The Three Evils” following the violence of July 2009118. During the 2008 Beijing Olympics, one anonymous Uyghur blogger explained the Uyghur fears that Han-ness was being objectivised as a national mode of self-identification:

Zhonghua Minzu takes Han as the nucleus and includes 56 internal minzu groups. Perhaps because the Han minzu have a huge population, the national language is Hanyu119, culture takes Han culture as its nucleus, so in reality we often don’t consider the religious faith or cultural customs of other minzu. We conflate Han minzu and Zhonghua Minzu, so that it is like Zhonghua Minzu and the Han minzu are the same and other minzu are not considered to exist. (Bense Shouyiren, 2008).

118 The site has since set up servers in Europe. See: http://www.uighurbiz.net/. 119 Hanyu refers to Mandarin Chinese (Zhongwen) but literally means “language of the Han” and its use by Uyghurs indicates its outside-ness. Hanzu tili is used in the same way in Uyghur language.

179

Echoes of Ma Rong’s (2010) argument that “Hanification is modernisation” were seen in official discourses of minzu tuanjie and sange libukai. The conflation between Han-ness and Chinese- ness was seen in the analysis of “bilingual education”, which frames Mandarin as transcendent of its Han Chinese origins and the only language of modernity in China (Ministry of Information, 2009, p.92). This objectivisation is being challenged by Uyghurs as a form of marginalisation of Uyghur-ness. Symbols of Han ethnicity become symbols of national self-identification so at the national level, no other minzu exist. The Uyghur blogger above exemplified these concerns amongst Uyghurs by saying “Han symbols” such as the “sons of the Yellow Emperor”, “the dragon”, and the “Great Wall” cannot be the symbols of Zhonghua Minzu. Echoing the sentiments repeated by dozens of informants, these “symbols and totems” “do not consider the feelings of other minzu” (Bense Shouyiren, 2008). Uyghurs, Tibetans, Mongolians, and Hui thus will feel they “do not belong to Zhonghua Minzu” (Bense Shouyiren, 2008). As part of the more recent debates on a “second generation” of ethnic policies, discussed in chapter 2, Zhu Weiqun, vice director of the United Front work department, acknowledged the existence of Han “splittist elements” and “big-nationality chauvinism” in the phrases “descendents of the dragon” and “Sons of the fiery emperor and the Yellow Emperor” (Zhu, 2012). A less-celebrated IR theorist and scholar of American Studies at the China Institute of Contemporary International Relations, Da Wei, asked “who are the Chinese people?” in his book of the same title (Da, 2010). His arguments are an implicit critique of Fei Xiaotong’s conceptualisation of China as Han-centred and indicate the potential for debates on identity politics outside Xinjiang. Da Wei argues that if China is to be a “pluralistic but unified” nation, it is not enough to merely be “introduced” to Chinese cultures other than that of the Central Plains (Da, 2010, p.6). He goes on to say that “genuine” cultural revival cannot be based solely on celebration of Confucianism and traditional festivals of the Central Plains and Chinese people should celebrate festivals of all groups in China (Da, 2010, p.5-6). However, Uyghur websites discussing the issue are closed down in a drive against “The Three Evils”. This criminalisation of speech marginalises Uyghurs from discussions on their own self-identifications, thereby continuing to produce a Han-centred nation. These symbols of Han ethnic identity remain in use to represent the entire Chinese nation at home and abroad.

The performance and representation of traditional festivals offer a window into the official model of Chinese nation-building. Through seemingly apolitical methods, the different meanings officially attributed to these festivals take Han-ness as a trans-ethnic category capable of conversion. The very public presence of celebrations of Han festivals contrasted against the relative absence of for Uyghur festivals. This performed China through the mutually constitutive national/ethnic and public/private dichotomies. However, these performances do not acknowledge that Han is an ethnic category. Han-ness is thus being objectivised as Chinese-ness and superimposed onto all other ethnic groups as a model of inclusion. The Han-

180 centred structure of China was formally performed at Xinjiang University on the 30th September, 2009. Staff formally distributed the “love Xinjiang, love China” stickers to each and every student in every class on the 30th September.120 This was accompanied by a brief explanation that it was important to celebrate national day and minzu tuanjie, assuming that unity and Chinese nationalism went hand in hand. These stickers were given out to every student along with a mooncake, traditionally eaten by Han for mid-autumn festival (Zhongqiu jie). The speeches by heads of departments claimed Mid-Autumn festival and Spring Festival (Chunjie) were China’s most traditional and most important festivals. The silencing of non-Han festivals and use of Han festivals as exemplars of the Chinese nation was a conflation of Han-ness and Chinese-ness. The social practices of the majority ethnic group came to represent an entire multi-ethnic nation. This objectivisation could also be seen in slogans raised across the university such as “Joyfully celebrate National Day, happily welcome Mid-Autumn Festival” (see figure 5.6). Engaging in social practices associated with the Han ethnic group is articulated through the celebration of national day, thus a means to become Chinese.

Figure 5.6: Top: “Joyfully celebrate National Day, happily welcome Mid-Autumn Festival”. Bottom: “Strengthen Minzu Tuanjie, protect social stability”.

Mid-Autumn festival was recognised as a national-level “intangible cultural heritage” in 2006 and became a national holiday in 2008 (Xinhua, 2011). The festival is celebrated on the 15th

120 This is based on observation and discussions with staff and students of the university.

181 day of the 8th month of the Chinese Lunar calendar. It is said to originate in moon worship from the Western Zhou period (1046-771 BCE). Officially, mooncakes are said to have first been made in the 14th century. “People” exchanged pancakes stuck with secret messages saying “Kill the Mongols on the 15th day of the 8th month” from rebel leader Zhu Yuanzhang to organise Chinese rebellion against the Mongols (Xinhua, 2011). The message at the university was that to love Xinjiang one must love China. This love required members of the nation to eat mooncakes in celebration of violent rebellion against non-Han seeking to rule China. This meant loving China as Han-centred. The regional government sent an SMS text to all residents of Xinjiang “wishing all minzu a joyous celebration of mid-autumn festival, a tradition of Zhonghua Minzu”. By celebrating this as a Chinese national celebration, the internal and hierarchical ethnic distinction between civilisation and barbarism, Hua and Yi, is performed as the ethnocentric life of the nation. The performance at the university represents a broader unconscious objectivisation of Han-ness as Chinese-ness. As Ma Rong tells us, non-Han ethnic groups are not qualitatively different; they are just behind the Han in terms of their levels of acculturation to a singular form of civilisation and modernity (Ma, 2007, p.5).

This festival, like all festivals officially presented as “traditional Chinese festivals”, is not celebrated by Uyghurs in Xinjiang, or Mongols for that matter121. This was observed in the city during Spring festival and Mid-Autumn festival. Uyghur areas were quiet but Han areas were bustling with people setting off firecrackers. Spring festival saw a public celebration at People’s Square with banners to celebrate the occasion and gatherings of Han families. In Uyghur areas the only observable difference to daily life was an increase in paramilitary patrols, including armed stations set up outside Mosques with the slogan “the Police are here to celebrate Spring Festival”. The objectivisation of Han-ness as Chinese-ness or demand for acculturation of non- Han groups is being resisted by Uyghurs through non-participation in these festivals. Most Uyghur informants said they “hated firecrackers” and “look forward to it being over”. One mid- 20s male student directly acknowledged the inequality by asking “why do they get to celebrate and we don’t? Why do I have to listen to their firecrackers over and over yet we aren’t even allowed to gather in public?”. After the festival in October 2008, the anonymous blogger, Uighur, published a familiar story to that told by many of my informants on the forum Uighurbiz. Uighur said he was approached aggressively by a cadre in his residential block asking him why he wasn’t taking a mooncake provided by the local government:

‘Aren’t you going to celebrate Mid-Autumn festival?’ I replied that I wasn’t. He very sharply looked at me ‘Are you not a Chinese person?’ I said, 'I am a Chinese person!’.

121 When interviewed, most Uyghurs would laugh at the very idea. I eventually stopped asking Uyghurs if they celebrated Mid-Autumn festival or Spring festival. Originally, I thought it could provoke interesting response. However, it offended some people who then ended the conversation assuming I knew nothing about Xinjiang.

182

He said: ‘Who do you think celebrates Mid-Autumn festival?’ I was a little tongue-tied but I said: ‘Han’. I know that in Xinjiang discussion of these types of topics is very sensitive. They can casually say Uyghurs are like this, Uyghurs are like that. But if we are in front of them it is very hard to say Han are like this, Han are like that because it is said to be equivalent to opposing them!! Afterwards he then said ‘Chinese people must celebrate Mid-Autumn festival’ and repeated this several times. Without any resistance I said ‘This should be the tradition’. He then didn’t pay attention to me but in my heart I felt so uncomfortable. We get along harmoniously with them and we protect the unity (tuanjie) of the nation. Yet why do we have to be subject to their culture, regardless of whether or not we have this kind of tradition ourselves? (Uighur, 2008)

Uighur was not objecting to harmony and unity per se. He was resisting the hierarchical nature of tuanjie where Han culture is unconsciously framed as superior to that of other ethnic groups inside Zhonghua Minzu. By framing Han social practices as Chinese and thus that which Uyghurs are to be converted to created resistance to nation-building because it means Hanification. If we understand self-identification following Cynthia Weber and Judith Butler as a “signifying practice”, resistance to hegemonic discourses is enabled by this discourse itself such that the choice is not whether to repeat them but how to repeat them (Butler, 1999, p.184-185). As Homi Bhabha tells us, the textual process of political antagonism, that is the rejection of the identifications offered by states, involves a “reading between the lines” (Bhabha, 2009, p.35). The producer of the text can become the inverted, projected object of the argument turned against itself (Bhabha, 2009, p.35). Here we see tuanjie being inverted and turned against itself, like we saw with claims to “justice” and “equality”. Tuanjie is inverted to performatively enact an equal social position in China, which Uyghurs and other ethnic minorities currently do not occupy. This is a “taking up of the tools where they lie, where the very ‘taking up’ is enabled by the tool lying there” (Butler, 1999, p.185). Another Uyghur blogger repeated the family metaphor discussed in the previous section to argue that to celebrate the Great Wall and the defeat of the Mongols will only create “hostilities” between “brothers” and that “we are all one family (Yi Jiaren)” (Bense Shouyiren, 2008).

Comparing these celebrations with those of Uyghur festivals can illuminate the different social positions of Han and Uyghurs in Xinjiang. In 19th century Xinjiang Qurban Heyti and Roza Heyti122 marked the beginning and end of a month of fasting for Ramadan. Qurban celebrates Abraham's willingness to sacrifice his son to God while Roza commemorates the revelation of Quran to the prophet. Roza and Qurban were regarded as the most important festivals by locals though precise interpretations over their meaning varied over time and from place to place (Bellér-Hann, 2007, p.350-355). Even today, in traditional Uyghur mähällä residential areas the month of Ramadan is a time of huge of cultural significance. These rhythms and routines punctuate daily life in ways which mark it off from other months (Dautcher, 2009, p.283). The rhythms and routines of fasting by day and eating by night shape work and social life as

122 In the Arabic speaking world these are referred to as Eid ul-Fitr and Eid al-Adha.

183 restaurants are closed during the day and crowded after dark. When I asked a 40 year old female Uyghur language teacher why Uyghurs use the word bahyrem for national day and mid- autumn festival but use heyti for Uyghur festivals, she responded that bahyrem was for Chinese national holidays where heyti was used for “our traditional holidays”. I then enquired why festivals, like Spring Festival (Chunjie), which have been celebrated by Han for centuries are not seen as “traditional”. The response was “yes, I know but heyti is for our holidays, not for Han holidays”. This was a (re)performance of the sharp Self/Other and ethnic/national distinction circulated in party-state discourse. It reflected the articulation of Chinese-ness as Han-ness but Othered China as an external culture rather than an attractive form of self-identification. Traditional Uyghur festivals are considerably less officially celebrated than Han Chinese festivals in Xinjiang and these comments reflect an awareness of this social context. This differential approach to Uyghur and Han festivals produces a marginalised social position for Uyghurs in China. Symbols of Han-ness become normalised as national Chinese symbols associated with the public sphere where those of Uyghurs are relegated to the private sphere and ethnicity. Heyti is then embraced by Uyghurs as a private, ethnic sphere of celebration, which is seen as outside the identity politics associated with bahyrem and Chinese nationalism.

Life in urban Ürümchi today is unsurprisingly different from 19th century Xinjiang. For example, urban intellectuals, who often veer away from religion and towards Turkic language as an identity marker, rarely choose to fast (Dautcher, 2009, p.285). I observed this in Uyghur restaurants across the city, which stayed open throughout the period, in contrast to those which remained closed during a visit to Kashgar for Ramadan 2007. Nevertheless, many students and teachers complained to me that they were prohibited from fasting. In 2012 international news reported the party-state’s drive to prevent Muslims in Xinjiang from fasting123. Government notices are posted reminding people to “eat properly for study and work” and large banquets are recommended as “gifts” for Muslim state-employees during the fasting period (The Guardian, 2012). Students at three separate universities complained to me that security guards monitored their dormitories to ensure they were not praying. Many students said they would secretly leave their dormitories to visit the Mosque whenever they could. Many teachers and staff at these universities explained that they would suddenly be invited out for many lunches during this period by officials at the school. These informants saw this monitoring of their behaviour as a form of social control to ensure that they did not fast. Some teachers did fast but had to do so in secret. One female language teacher in her mid-30s explained to me it was “hard enough” to fast without “all the Han and party-members trying to make me eat and

123 Restrictions on religious practice in Xinjiang have only intensified since this fieldwork. For example, in , 12 children were injured in a raid by police on a religious school (The Independent, 2012). Official notices of mandatory of house-searches in Uyghur neighbourhoods for “illegal religious materials” and bans on “religious clothing” in commercial shopping areas are being regularly displayed (UHRP 2012a; 2012b; 2012c).

184 being nosy. I just tell them I’m sick and can’t eat”. The textual environment of Ürümchi, where “the inside/outside Three Evils” were a constant “threat” did not change the religious origins behind fasting. However, the party-state’s identity politics did shape how Uyghurs imbued religion with meanings which involved secretive resistance against Hanification. Given the political restrictions on practicing Islam and in undertaking this type of research, it would be impossible to accurately gauge the level of importance Uyghurs attach to religion. Nevertheless, it is important to highlight the different ways of being Uyghur in Ürümchi. It is also likely that party-state restrictions on religion in Xinjiang will encourage Uyghurs to turn towards Islam and against China. For example, the teacher quoted above told me in 2008 that fasting was not compatible with modern work patterns. However, by 2012 she strictly participated in fasting and explained it was an important part of “being a Muslim and maintaining our culture”.

Han festivals were an important part of public and thus, national life where Uyghur festivals were a private and ‘ethnic’ affair. For example, Universities did not celebrate Uyghurs festivals. Roza fell on the 21st September in 2009. This was just a week before Mid-Autumn Festival but it passed without official mention or occasion. This marginalised the position of Uyghurs and other Muslims in Xinjiang as domesticated and ethnic and outside the public, national sphere. The white paper, Development and Progress in Xinjiang, was printed in full in the local Chen Bao newspaper on the 22nd stating “minzu who follow Islam warmly celebrate Roza and Qurban, Han, Mongolians, and other minzu celebrate Spring festival” (State Council, 2009b, p.39). By not listing Uyghurs as welcoming spring Festival, this was tacit acknowledgement that they do not. However, by stating “other minzu” do, the implication was that Han-ness was trans-ethnic rather than specific to one ethnic or religious category. In contrast to the national seven-day holiday for spring Festival, the Regional Government posted an official notification for Roza stating that Muslims could have one day off while others had to go to work (Chen Bao, 2009s). Nevertheless, Muslims in headscarves who worked for the local government and in universities were observed across the city cleaning streets and toilets on that day. City-life and the self- conscious identities of Uyghur intellectuals shaped these celebrations. However, power relations between the state and Uyghurs also played an influential role. Religious restrictions meant those who work for the state had no choice but to abstain from celebrations until they finished work for the day. Xinjiang Television news broadcasts and SMS texts from the Regional government had invited “every minzu” to “celebrate” Mid-Autumn and Spring Festivals. However, these same organisations asked those “minzu who practice Islam” to “harmoniously celebrate” Roza. This was a warning which highlighted the “threat” of the ethnic with cautionary reminders to be harmonious against the security of the national.

The party-state de-Islamicises these traditions by omitting the role of Islam in the public narrative. The minzu tuanjie education texts analysed here and news reports on the subject

185 give no explanation of the significance of these festivals or fasting. This silencing socially marginalises these festivals vis-à-vis those of the Han because their meaning is merely noted as Islamic or shaoshu minzu in these texts. The front page of the Chen Bao on the 22nd September showed an old Uyghur lady celebrating by presenting SWAT police with home-made pilav124 alongside a raft of extracts from the Progress and Development of Xinjiang (Chen Bao, 2009t). Emphasising unity with the party-state through the gift of food from Uyghurs to the security apparatus failed to acknowledge the social and religious significance of these celebrations. This silenced Uyghur perspectives on their most important events in the social calendar and gave no voice to the religious meanings they articulate through them. The square around the main tower at Erdaoqiao/Döngköwrük saw hundreds of Uyghurs dancing together during the afternoon. This was semi-choreographed by a professional dancer but passers-by joined in and danced in different ways. This became a centre-less party without fixed meaning but through which Uyghurs could celebrate the festival in different ways. The celebration was what Anthony Cohen called an “expression of self-identity” through its re-appropriation for the requirements and propensities of the individual (Cohen, 1996, p.805). Participation could mean different things for different individuals but as a symbol of the group it retained its power to unify them in a single act (Cohen, 1996, p.805; 1985, p.98).

No Han Chinese residents were present at the Roza dancing. This reflected and reinforced the discourse that this was a non-Chinese festival for Muslims only. By late afternoon the square was closed and guarded by armed paramilitary forces. Increased patrols could be seen throughout the city including trucks with loudspeakers asking residents to “build a harmonious society”, again marking the day as particularly dangerous as Uyghurs gathered to celebrate. The male university student in his early twenties, who “hated the firecrackers” of Spring Festival explained he was “sad that Roza was disappearing” and that “it used to be such a long party where everyone could stay out all night but now we can’t do anything, just walk around”. The celebrations were shaped by ethnicised security practices. These practices performed a Han- centred nation by socially ordering the meaning of different festivals and different minzu. The teacher who secretly fasted said “I don’t know how they can celebrate at this time, in this terrible situation”. She later explained the political restrictions and disappearances after July 2009 meant that life was unbearable and there was no purpose to celebrating unless it could be done freely. Identity politics of the Chinese party-state, which drew from Han nationalist discourses, was subsequently shaping how Uyghurs understood performances of their own self- identification.

124 Polo is the Uyghur take on Pilav, served all across Central Asia. This dish of rice, mutton, carrots, and raisins is one of the staples of the Uyghur diet. However, it is usually seen as a luxury compared to the more everyday noodles (laghman) and is served at important social ceremonies.

186

The micro-fieldwork approach employed here should not be seen as separate from the broader study of Chinese politics and foreign policy. As discussed in section 2 of this chapter, the domestic imperative of minzu tuanjie is represented as the means to strength and “soft power” at the global level. The “soft power” cottage industry in studies of Chinese IR and foreign policy is focused as much on how to make China attractive to those outside China as it is to promote national cohesion at the domestic level (Young and Jong, 2008, p.459). China’s foreign policy discourse and the policing of domestic identities are inter-related practices of the articulation and security of Chinese-ness. Debates on the “China model” of economic development as an alternative to “Western arrogance” are framed as based on the “unique” “Chinese outlook” (Pan, 2007, p.3). This discourse is framed as a challenge to “western” power and the capacity to define “the West” will only grow with China’s international rise. However, the “China model” stems from a position of domestic power to articulate what it means to be Chinese in the first- place. Da Wei urged Chinese people to celebrate all ethnic festivals if China was to be a multi- ethnic nation. However, this did not occur in Ürümchi. In fact the party-state produced a hierarchical and Han-centred China through these celebrations. This ethnocentric model speaks to Han nationalism because it prioritises, celebrates, and ranks symbols of Han ethnicity over those of all other ethnic groups. The party-state positions itself as a neutral representative of minzu tuanjie. However, Han-ness is unconsciously understood as non-ethnic. Han and Uyghurs then (re)perform ethnic boundaries when they celebrated separate ethnic festivals in different ways.

The stories written by Uyghur bloggers and the differential celebration of “ethnic” festivals reflect an alienation from Han-ness objectivised as Chinese-ness. Uyghurs tend to understand themselves as ethnically and culturally distinct from Han traditions and not as Ma Rong claims, the same but behind. Da Wei is seeking to “rectify nationalism” to include ethnic minorities and make Chinese people ask if the dragon and the yellow emperor, symbols of the central plains, can legitimately represent a multi-ethnic nation (Da, 2010, p.6 & 44). However, Zhonghua Minzu in Xinjiang is closer to that imagined by Fei Xiaotong and Ma Rong. It centres symbols associated with Han-ness and represents them as modernity. This then offers Uyghurs and other ethnic minorities the opportunity to be converted from barbarians (Yi) to civilisation (Hua). However, as we will see in chapter 6, Uyghurs are not being absorbed or attracted by the Han nucleus contrary to Fei’s claims of inevitability. Uyghurs generally choose not to participate in performances of Chinese nationalism because they understand it as a form of subjection to Han-ness rather than inclusion in a multi-ethnic national community.

187

Conclusions

This chapter showed how official Chinese and unofficial Han nationalism in Xinjiang shape each other in processes of identity contestation. These contestations have created a marginalised social position for Uyghurs in the party-state’s nation-building project. Through minzu tuanjie, this project places the Han at the centre by objectivising social practices associated with ethnic Han as national and inclusive.

The first section focused on the inclusion of Uyghurs in Chinese nation-building through the public performances of minzu tuanjie and Zhonghua Minzu. This showed how the inclusion of minzu tuanjie and the exclusion of “The Three Evils” are inter-textually related and mutually constitutive aspects of the same nation-building “struggle”. Uyghurs are being included through minzu tuanjie in a Han-centred nation, which places Uyghurs in a position of cultural and economic backwardness. They are to be converted into nationals but the symbols used to define the nation are understood by Uyghurs and Han alike as symbols of Han ethnicity (e.g. Mandarin Chinese, the Great Wall, and the Yellow Emperor). This indicates how the official model of nation-building is being shaped by the ethnocentrism of Han nationalism. Symbols of Han-ness are being objectivised as national symbols with the potential to convert Uyghurs. The final section showed how the representation and celebration of different traditional festivals in Ürümchi are an example of this objectivisation. Uyghur Festivals were represented as ethnic in contrast to national Han festivals. These official representations order categories of ethnicity through the prism of the ethnic/national dichotomy.

Discourses on the unbroken, timelessness of Chinese culture have resonance for some Han. They maintain the appearance of cultural continuity and offer a model of belonging in the face of the modernity which transforms and re-shapes community. It is less fruitful to analyse modernity and tradition and official and unofficial in terms of their essential theoretical foundations. Instead, this chapter explored how the often logically incompatible foundational commitments of tradition and modernity and the official and the popular as social practices can actually be mutually reinforcing. The cultural inclusion/conversion of Confucianism, the modernising imperatives of the Leninist minzu project, and the much-ignored persistence of “great Han chauvinism” are all based on competing views of human communities. However, in practice they are bound together in ways which illuminate the instability and socially contextualised nature of the self. Logically incompatible “foundations” of community can exist together without disturbing their ontological premises. The real power of nationalism is that individuals can emphasise different discourses on the nation for their own ends depending on social context and which Other is to be excluded. These selections often take place without disturbing what may appear to be contradictions or theoretical inconsistencies (“multi-ethnic

188 nation” vs. the “Han nucleus”). As Benedict Anderson wrote, one of the most vexing aspects of theorising nationalism is its ongoing political power, which contrasts against its “philosophical poverty” (Anderson, 1991, p.5). Often we need to “break with common sense”, such as the seemingly uncontroversial point that ethnic or national conflicts inherently involve conflict between ethnic groups or nations (Brubaker, 2002, p.166). Instead, we paid attention here to how conflict is ethnicised and how group-ness is attributed to individuals.

Nationalism neither comes from above nor below. The official and unofficial interact and shape one another in a dialogical relationship. The national identity offered by the party-state does not cause popular Han nationalism but shapes and is shaped by it. The official articulations of Han-ness are not reflective of a pre-existing fixed identity. They offer a discursive category of identification, which people contest and re-negotiate. The party-state does not directly address Han nationalism. This allows Han nationalism to remain safe and to be equated with Chinese- ness unlike , which is subject to wide ranging, public securitisations as a threat. Chinese nation-building in Xinjiang is inherently exclusionary and hierarchically inclusionary. Unconscious prioritisation of Han-ness and the maintenance of ethnic boundaries in Xinjiang are akin to the “unconscious hanging of the flag” within Billig’s idea of banal nationalism (Billig, 1995, p.6). They are taken for granted ideologies but rather than articulating externally bounded nations or a unity of equals, this is nationalism based on ethnic separation and internal differentiation. Ma Rong argues that Chinese civilisation, and indeed “bilingual education” is based on “teaching without discrimination” such that “barbarians” can become Chinese through acculturation (jiaohua) (Ma, 2007, p.7). However, in practice, political life in Xinjiang is based on a structural framework of Minzu. Different ethnic groups are separated in education and through public performance into the mutually constitutive categories of Han and Shaoshu Minzu discussed in chapter 2. This inclusion demands identification with their position as less civilised and less developed than the category of Han. For example, Uyghurs must accept that Zhonghua Minzu is a social hierarchy where the Mandarin language represents civilisation (Hua) and the languages of ethnic minorities represent barbarism (Yi).

In some ways, Tuanjie in Xinjiang is formalistic. It does not address real life problems of discrimination and division. However, it cannot be dismissed as mere propaganda. Minzu tuanjie provides the framework to so much public discourse which people consume and contest. It informs the overall approach to nation-building policies in Xinjiang. The party-state’s concept of “ethnic extinction” (minzu xiaowang), discussed in chapter 2, acknowledges that tuanjie is a temporary ideology until modernity washes away ethnicity. However, the objectivisation of ethnic Han-ness further contributes to feelings of exclusion, assimilation, and insecurity on behalf of Uyghurs. Tuanjie’s objectivisation transgresses existing boundaries of community and minzu in ways which create violent conflict over social positions in China. Nation-building and

189 ethnic relations are inseparable in Xinjiang because the state positions itself between two ethnic groups, which identify themselves as distinct. However, this positioning is far from ‘neutral’. The party-state promotes ‘unity’ and security in ways which exacerbate conflict. Securitisation makes Uyghurs feel insecure because it unconsciously prioritises its understanding of Han-ness over the self-identifications of Uyghurs. The remaining chapters will now turn to the ‘unofficial’ perspectives of Han and Uyghurs in Ürümchi on ethnicity and nation through the use of detailed, semi-structured interviews.

190

Chapter 6: (Re)Performing the Inclusion/Exclusion of Uyghur-ness, (Re)Performing China

Introduction

The previous chapter showed how inclusion and conversion (minzu tuanjie) on the one hand and exclusion and securitisation (“The Three Evils”) are mutually constitutive aspects of nation- building in Xinjiang. As this chapter will show, the characteristics through which Uyghurs tend to self-identity (language and religion) are those the party-state seeks to exclude as un-Chinese. The focus in this chapter is how Han and Uyghurs define the Self and the Other through one another. Self/Other-identification is a way of acknowledging individuals identify themselves but do so by excluding other individuals from their sense of we-ness. The methodological approach in the final two chapters shifts away from directly analysing articulations of identity in the official discourse of the party-state from above. The analysis now centres on self-identifications from below through interviews with Han and Uyghur residents of Ürümchi. This will evaluate the ‘effectiveness’ of nation-building to explore how the identity politics from above is received by those it seeks to identify in their social practices and their private self-identifications. Does the purported nation understand itself through the boundaries offered by official nationalism? The separation of above and below is primarily for presentational and methodological convenience. Official and unofficial should not be placed in states of ontological quarantine. The ontological approach here analyses the official and the unofficial through the relations of post- structuralism rather than the billiard ball causality of Positivism. Instead of adopting the language and epistemology of cause and effect, the analysis seeks to explore how the official and unofficial mutually shape one another through the concept of (re)performance.

(Re)performance is used to conceptualise the response of the unofficial to the official, which permits the examination of purported nations and not just nationalists. The suffix (re) methodologically separates the official party-state articulations of identity-security from unofficial Han and Uyghur responses. It would be normatively and ontologically problematic to assume the party-state speaks for all or even most people whom it defines as Chinese. This would only reinforce what Gayatri Spivak (1988) called “epistemic violence” through unproblematised categories of identity which emerge and are imposed from above through power relations. “Representation has not withered away” (Spivak, 1988, p.104). Representation will not wither away if some groups have more power than others to articulate identities for

191 people and for these articulations to be heard. (Re)performance explores how official discourses on identity are inverted against themselves by those they seek to identify.

This chapter will use Bhabha’s idea of “reading between the lines” (2009) to show how Uyghur- ness and Chinese-ness are being (re)performed. Hegemonic discourses are inter-textually inverted as modes of resistance. Inter-textual performances of identity by Uyghurs and Han challenge and reconfigure the meanings of official categories of modernity, nation, and ethnicity. As Edward Said explained, “contrapuntal reading must take account of both process, that of imperialism and that of resistance to it”, such that the goal is to reveal the wholeness of the text and the overlapping, mutually embedded histories of metropolitan and colonised societies and of the elite and subaltern (Said, 1993, p.66-67). We thereby must understand the social forces from above and below which shape discourses on nation and resistance to understand the “wholeness” of nation-building. “Wholeness” does not refer to nation-building as a totality or a coherent singular discourse. “Wholeness” highlights the tensions and inter-connections between official and unofficial politics and allows us to examine the mutual constitution of official and unofficial and of elite and subaltern in the production of political community.

Following Vered Amit’s critique of Benedict Anderson’s “imagined political community” (1991), community is not simply conceptual but also visceral. Communities are constructed through social practices amongst community members, who unlike in an imagined community do in fact know one another (Amit, 2002, p.18). The boundaries articulated by states can be politically effective but their meaning only truly be felt by potential members if they are able to realise them in social relations which are symbolically close to them (Amit, 2002, p.8). Nation-building attempts to produce community and its boundaries require that these attempts provide meaning and belonging-ness. This belonging-ness must have some sense of correspondence with its postulated members’ lived experiences and how they think about them to be effective (Cohen, 1985, p.13 & 28). Community is constructed symbolically and exists in the minds of its members such that the ‘doing’ community matters less than the ‘thinking’ about what its ‘doing’ means (Cohen, 1985, p.98). Communities defined through ethnicity may feel more visceral to their members because they are based on seemingly material bonds such as language use or religious practice; a shared form of knowledge of how to do things. However, while material and visceral social practices may have their own conditions and effects, they have no meaning without discourse (Hall, 1996, p.444). Viscerality then is as much a discourse as the imagination of national belonging, even if it as very different one.

The methodology in these final two chapters employs semi-structured detailed interviews with Han and Uyghur residents of Ürümchi to explore their perspectives on nation-building and their

192 own Self/Other-identification 125. The first section will explore how Han (re)perform identity- security discourses in Ürümchi. The specific empirical concern here is how they employ pre- Communist discourses of lineage and ethnic Otherness through the division between Han and Inner China (neidi) on the one hand and Uyghurs and frontier (bianjiang) on the other. These boundaries are (re)performed through official discourses of “liberation” where the Han are not simply the nucleus of the nation, they are the nation. The second section examines Uyghur (re)performances of identity-security. This will explore how Uyghurs draw from pre-1949 discourses of imagined community (Turk and Islam) to position themselves outside Zhonghua Minzu through reference to daily, visceral experiences. This section analyses how daily practices are imbued with identity politics by Uyghurs, which are shaped by but challenge discourses of Zhonghua Minzu. This chapter examine the boundaries of inclusion and exclusion, which Han and Uyghurs perform in their self-selected categories of Self/Other-identification. The argument presented here is that both Uyghurs and Han in Ürümchi tend to view themselves as distinct communities through self-articulations of ethnicised viscerality. These Self/Other-identifications are contrasted by these communities against the multi-ethnic nationhood and conceptual imagination of minzu tuanjie thus Othering the CCP as outside national community. Identity politics are being imbued with viscerality to perform ethnic origins and discontent with the current social order.

6.1: (Re)Performing Han-ness, (Re)Performing Inclusion / Exclusion

This section will examine how Han in Ürümchi frame their self-identifications by (re)performing official conceptualisations of identity-security. The section will ask how do these (re)performances correspond or conflict with the official model of Chinese nation-building in Xinjiang? The analysis will proceed through three sub-sections of lineage, religion and food, and social contact with Uyghurs. It will explore the contentions of Leibold (2010) and Gries (2004) that the party-state has lost control of the monopoly of Chinese nationalist representations of Self and Other, particularly since the onset of opening and reform in 1978. The discussion of small-scale assaults in Ürümchi in late 2009 in chapter 4 showed how the historicity of the category of Han is concealed in contemporary China. Han-ness was objectivised as Chinese- ness where Uyghur-ness is Othered as a source of infection to the national geo-body. The previous chapter outlined the historical development of the minzu category. Kai-Wing Chow’s (2001) framework of “semantic hybridity” illuminates how European and Japanese discourses of “race” became absorbed into Confucian universalism, reshaping each other rather than causing cultural rupture or revolutionary transformation. If we use semantic hybridity rather than revolution to frame social and political change, we can see how Otherness was reconfigured

125 Chapter 1, section 4 fully outlined the methodology employed in these chapters.

193 from hua/yi to race (zhongzu) to minzu maintaining non-Han groups as outsiders. This section will explore the persisting importance of discourses of lineage in contemporary China and how these are adapted to contemporary social practices in the framings of Self/Other-identification by Han in Ürümchi.

The modern Han ethnonym builds on older historical formations, which emerged through complex interaction and distinction between the sedentary dwellers of the central plains and of the northern steppe (Leibold, 2010, p.10). The historical and contemporary usage of the term is highly unstable but its origins emerge with the intervention of the Western Hu people who labelled the sedentary dwellers of the Central Plains as Han despite the fact that they were politically and culturally divided for much of their history (Elliot, 2011, p.173-175). Older discourses which divide civilisation from barbarism and nomadic from sedentary peoples inform the symbols and boundaries that define the Han today (Leibold, 2010, p.10). Symbols such as the dragon and boundaries such as the Great Wall represent divisions between sedentary dwellers of the Central Plains (Hua) and nomadic northerners (Yi). These boundaries have been drawn through this interaction and remain indicative of the tension between the Han as an ethnic group and the Han as the trans-ethnic leaders of all ethnic groups in China. The fluidity of Han as an ethnic and cultural marker reflects the tension between two competing ideologies of political community based on inclusive cultural universalism and exclusive ethnocentrism (Leibold, 2010, p.10). The former seeks to convert or assimilate non-Han groups, where the latter excludes them from the majority ethnic group, understood as the nucleus of the nation. Disentangling these two discourses helps us to build theoretical models of inclusion/exclusion in China. However, social practices in Xinjiang are often understood through both of these formulations of Han-ness by different individuals at the same time. As we saw in Chapters 3 and 4, exclusion awaits those Uyghurs who resist conversion and discourses of “The Three Evils” and inconvertibility emerges. In the contemporary context, inclusion and exclusion are mutually reinforcing ideas. The civilising project and contemporary nation-building may appear to be based on competing and incompatible ontologies of human community (pre- modern, inclusive culturalism VS modern, exclusive nationalism). However, in practice they are mutually reinforcing discourses (Callahan, 2010, p.131).

6.1.1: Lineage and Self-identification Amongst Han in Ürümchi

Mrs Du126, in her late 30s, was born and raised in Xinjiang. She maintained connections with Shanghai where her parents had returned after spending the 1950s and 1960s in Xinjiang, in her words, “opening the frontier”. Mrs Du represents the Ürümchi Han intellectual as a

126 All names used in this chapter and all other interviews in the thesis have been altered and used for convenience only.

194 professional university teacher. Her marriage to a modestly wealthy businessman offered her luxuries such as travel opportunities across China not afforded by many other interviewees. Mrs Du was selected here because her positive views on minzu tuanjie coincided with other Han intellectuals interviewed. Many intellectuals in Xinjiang echo the party-state’s discourse on minzu tuanjie. For example, Miss Lan, a mid-20s female Mandarin Chinese teacher from the countryside north of Ürümchi, explained to me that “minzu tuanjie is an ancient tradition of the Chinese people”. One mid-40s male head of a university department, Mr Gu, often repeated the slogan “minzu tuanjie is good”. Mrs Du could communicate in basic Uyghur which is uncommon amongst Han. She expressed pride that this allowed her to “understand ethnic minorities”. Du explained she was tired having to justify this to her friends who would say it’s a “useless” or “backward” language. This respect for diversity was not shared amongst all social classes. For example, in a typical answer to the question “can you speak Uyghur?”, one 45 year old male taxi driver from Xi’an, explained that “this is China (Zhongguo), everyone should speak Chinese (Hanyu)”. This elided the difference between Chinese and Han and objectivised Han-ness (Hanyu) as Chinese-ness (Zhongguo). Amy Kardos’ (2010) study of self-identification among Han settlers in , a famous oil-town in the north of Xinjiang, found that “Xinjiang Han” displayed tolerance of difference and ethnic diversity. However, Karamay is a rich oil-town with a GDP per capita higher than most states in Western Europe. Kardos’ sample, as well as those interviewed from the intellectual class here, should be viewed as representative of a privileged sector of Xinjiang society with quite different perspectives from other social classes on identity politics.

Mrs Du advised me to read kouhao because they “tell you how society works”127. This as we shall see differs from the popular rejection of meanings behind slogans in Ürümchi. Mrs Du was relatively politically engaged. However, she displayed critical reflexivity with regards to the identity categories promoted by the state, such as “Xinjiang person” (Xinjiang ren) and enjoyed discussing them. Jay Dautcher analysed the official interpellations of Uyghur identity through the concept of “Xinjiang person” in party-state slogans (Dautcher, 2009, p.40). The category articulates Han and Uyghurs as belonging to the same unit known as Xinjiang. Han are then intimately tied to a shared, conceptual community with ethnic minorities through minzu tuanjie. Fei Xiaotong’s (1988) classic thesis on the “unified, pluralistic” Chinese nation centred the internal boundary between hua/yi and between sedentary/nomadic cultures. Attachment to land and lineage are seen as the source of Chinese modernity through agricultural development128. However, these attachments to land and lineage run counter the idea of Xinjiang person, which as we shall see frames identity as ‘modern’ and multi-ethnic. Mrs Du, Miss Lan, and Mr Gu all

127 Mrs Du listed those analysed in chapters 4 and 5 such as sange libukai and minzu tuanjie slogans. 128 This binarism is repeated in all the minzu tuanjie education books and cadre training manuals analysed in chapters 2-5. For example, see: Ethnic Affairs Commission (2009); Ministry of Information (2009); Ethnic Unity Education Board (2009).

195 repeated the same discourses that for Han “land is very important”, particularly owning and working the land. During one informal interview I enquired Mrs Du why very few Han would describe themselves as a Xinjiang person. When asked if they were a Xinjiang person, responses were usually a denial such as “No, I am a Chinese person” or “no, I am Hanzu” from half a dozen getihu129. Mrs Du’s response illuminate how intellectual Han in Xinjiang engage with the identity categories made available to them by the party-state:

“This is a real problem. People here come from all over China but they work for a while and then they leave. For Han people, land (tudi) and their relationship with it is so important. So your ‘hometown’ (laojia130) is very important. You don’t just come from somewhere by living there for a while. Your family has to live there for generations. My family all worked here for decades to open the frontier. Then they retired and went back to Shanghai. But I define myself as a Xinjiang person. Because I was born here, I won’t leave when I retire. But it is a real problem because people don’t stay here and they don’t become Xinjiang people. This is only in Xinjiang because people come from all over and they don’t stay long. ‘Xinjiang person’ is a different type of category altogether. It is different from saying you are a ‘Beijing person’ or a ‘Shanghai person’, it has a different conceptual meaning.”

Mrs Du chose to understand herself as a Xinjiang person and lamented that Han left Xinjiang because they remained or even became attached to their laojia. Han can become Xinjiang people. Unlike laojia it is a contingent and self-reflexive category which does not refer to ancestral lineage but to birthplace and or self-identification131. This, as Mrs Du explained, is unlike the category of Beijing person which requires one’s ancestral lineage is located in Beijing. This understands Xinjiang person as a different type of category from other seemingly parallel local laojia identities in China. Xinjiang person invokes the party-state’s claim, discussed in chapter 2, that Xinjiang has since ancient times been a multi-cultural, frontier transit point. Xinjiang, thus defined, is framed through transitory migrations rather than what Jay Dautcher (2009) termed the “chthonic identity” of Uyghurs and the framing of the self through birthplace and land. Dautcher showed how the “Xinjiang person” category seeks to “desettle” the “chthonic identity” of Uyghurs where they are psychologically relocated to a multi-ethnic Xinjiang rather than the local community which they identify as crucial to their identity (Dautcher, 2009, p.50-59). The very naming of Xinjiang is a form of “desettlement” because it

129 Getihu is a social class of self-employed, small-scale entrepreneurs. This emerged with the dismantling of some industrial state-owned enterprises and the loosening of state controls on participation in the market since 1978. 130 Laojia is translated into English as ‘hometown’. However, it is better to leave it un-translated here because in practice its use includes reference to lineage which the English term does not necessarily encompass. 131 Bendi ren was used by many interviewees. The term is generally used to mean a local or native person. However, these interviews found that many people would say they are locals but then reject the category Xinjiang person and identify with their laojia in neidi.

196 silences East Turkestan as a category of identity. However, this relocation under “Xinjiang person” also involves desettlement of Han self-identification, framed through attachment to lineage and to land, because it seeks to supplant lineage with multi-ethnic nationalism.

Mrs Du, like the party-state, wanted Han to become Xinjiang people and thus part of a shared community with other ethnic groups. She also contested internal boundaries amongst Xinjiang Han through how the external boundary with other minzu was framed. This contestation of what it means to be Han was shown through her lament of the division between those who identified with being part of a multi-ethnic community and those who defined themselves solely through lineage and ethnic exclusivity. Mrs Du positively stereotyped Xinjiang people as “warm- hearted” and “more accepting of difference”. This was apparently due to both their diverse origins from across China and their experiences of living with different ethnic groups in Xinjiang. This draws from the Uyghur self-stereotyping as “hospitable”132 (mehman dost) used as a form of self-definition by many Uyghur interviewees of all classes. Miss Lan similarly described an “authentic” (didao) Xinjiang person as one who was born in Xinjiang, eats mutton, and treats people warmly. Miss Lan also said “I am a Xinjiang person” when I asked where she was from. She explained that the reason so few Han described themselves as Xinjiang persons was “because before 1949 there were hardly any Han here…this is a new period in history”. Miss Lan qualified her statements by saying artefacts and ruins on display at the regional museum show that there have always been Han in Xinjiang. This awareness of demographic transformation indicates how intellectual Han understand nation-building as a transformative process in Xinjiang, which involves at least as much change as continuity.

These interviews correspond with the findings of Amy Kardos (2010) that there is a discourse of commitment to diversity amongst Xinjiang Han, at least amongst elites and the privileged. When I asked Mrs Du why there was so much animosity towards Uyghurs, she fore-grounded the boundaries of class internal to the Han category, saying, “that’s just taxi drivers! This is a problem too, they just say anything, and it’s always very extreme, it isn’t objective”. Mrs Du was dismissing the perspective of lower class Han as subjective in contrast to her self-defined superior, “objective” position. Mrs Du was asserting a position of relative power vis à vis the working class to define what it meant to be Chinese. She used party-state discourse on minzu tuanjie to assert her power to identify. The power to identify conceals the heterogeneous nature of social categories. Meaning is given to these categories through what Homi Bhabha called the “struggle of identifications” to gain the power to identify (Bhabha, 2009, p.42-43). Han cannot be considered subaltern in Spivak’s (1988) sense of being “unable to speak”. However, differentiation within the category shows how different social classes within the category have different levels of power to be listened to. There are centres within peripheries

132 See Dautcher, Jay (2009) and Bellér-Hann, Ildikó (2008), p.202-203.

197 and peripheries within centres. Relations between subject-positions are constituted in Hershatter’s (1993) framework of “nested subalternity”. These relations illustrate the possibility of “multiple, relational degrees of subalternity” where some groups “distinguish themselves from and speak for those ‘below’, while allying themselves with and speaking to those above” (Hershatter, 1993, p.111). There are power relations within ethnic groups, ordinarily understood as majority or dominant, which exclude certain perspectives on what being a member means, which often cross over with discourses on inter-ethnic boundaries. Individuals invert hegemonic texts against themselves and maintain positions as speakers of Han-ness whilst repeating and mediating the identity politics and social practices of the party-state.

During this fieldwork, intellectuals were amongst the very few who understood themselves as part of the imagined community of minzu tuanjie. When asked “where are you from?” almost all working class and getihu Han interviewees stated firm identification with their laojia and not Xinjiang. Even many born and bred in Xinjiang continued to identify with their laojia in other parts of China. One such interviewee, a male getihu in his late 40s, explained “I am from here but my laojia is in ”. This exemplified Xinjiang’s ambivalent position in representations of ‘Chinese’ history by understanding himself as in Xinjiang but not yet of it. Another late 40s male getihu explained where he was from through ethnicity, stating “I am a Hanzu”. When asked if he was a Xinjiang person he stated that “just now I am from here but my laojia is in ”. He later explained, like many interviewees, that he had never visited his laojia and that his grandparents came to “open up Xinjiang”. During every interview with Han the issue of laojia arose with the vast majority saying they were born in Xinjiang but their laojia is in inner China (neidi133). Those getihu who identified as Xinjiang persons generally self-stereotyped the region in negative terms, highlighting Han-Uyghur ethnic divisions and “backwardness”. For example, Mr Hu, a mid-20s male taxi driver, proudly explained that he was a Xinjiang person. However, he drew community boundaries by Othering Uyghurs through the convergence of class and ethnicity: “it’s still very backward, because of the Uyghurs, we are all Chinese but they are a disagreeable and repugnant minzu”. One male 40 year old taxi driver, Mr Yan, who described himself as bendi ren, whose parents came from Shandong to “open the frontier” explained he “hated” Xinjiang, particularly Ürümchi because of “Uyghurs”: “They are a backward, stupid minzu, they are very dangerous, you can’t trust them…99% are all bad, their ideology, their way of thinking, is just totally different from Hanzu”. Differences between laojia were framed as safe but ethnic difference was represented as a danger. Community is being imagined through the prism of ethnicity by Han in Xinjiang.

133 This concept of inner China is used in Xinjiang to refer to the Han-inhabited central plains. This is distinct from Xinjiang, Tibet, and Inner Mongolia, known as the frontier (bianjiang).

198

Imagining laojia as central to Han as a minzu is a (re)performance of the ethno-taxonomy of the party-state but using pre-1949 framings of community as lineage. This employs the identity vocabulary of traditional and modern China to contest what it means to be Chinese. Mr Qiang, a 36 year old getihu, came to Xinjiang ten years ago to make money because making money was “impossible” in his laojia in the highlands of . He planned to return to his laojia when he had enough money to retire, which is representative of the transitory nature of the Xinjiang experience for many Han migrants. When asked where he was from, Mr Qiang told me:

“I am Han, a Han of China. I don’t like Uyghurs. They are a fighting-killing people. They are an offensive and savage people. They all want independence from China. They aren’t a Chinese minzu. They aren’t Chinese”.

“Why aren’t they Chinese?”

“Because of their language, they don’t speak Chinese. They are more like Indians or a Pakistani minzu”,

“They are all learning Chinese now. So what happens after that, what will be the result?”

“They still won’t be real Chinese. They all want independence but it’s simply not possible. It’s like the Soviet Union. When a country breaks up it becomes poor and weak and other countries interfere and attack it- so we need unity (tuanjie)”.

Mr Qiang drew boundaries between his community and Uyghurs by adopting the language of colonialist superiority and representing Uyghurs as an obstacle to China’s national strength and prosperity. He called Uyghurs “savages”, which was justified through their interpellation as foreign outsiders who cannot master the national language. Mr Qiang and other Hanzu would identify themselves differently in other contexts. In Ürümchi when speaking to a foreign outsider, he became a Hanzu eager to emphasise the internal unity of Han and their disunity with the culturally non-Chinese, Uyghur Other.

While terrorism and tuanjie dominate public discourse in Xinjiang, these are not solely for Uyghur consumption and conversion. The party-state seeks to transform the self-understanding of the Han in Xinjiang to be of Xinjiang rather than just in it. This impulse to convert Han into Xinjiang people runs counter to the popular Han self-understanding as intimately connected to their ancestral heritage as a site of their identity. Anthony Cohen’s argument (1985) that communities respond assertively to encroachment on their boundaries equally applies to attempts to Sinicise Uyghurs as it does to the desettlement of Han lineage and ethnicity. Discourses of lineage pre-date CCP rule as Chow (2001) showed and the party-state has largely failed to supplant these with a multi-ethnic nation despite decades of campaigns against Confucianism and tradition. The multi-ethnic model of China offered by the party-state may place Han at the centre. However, it runs counter to the visceral experiences of the many Han

199 in Xinjiang who stay temporarily for employment but continue to view the place as outside themselves. This was a problem as Mrs Du explained because it threatened the multicultural nature of Xinjiang of which the Han are an important part. However, elites and intellectuals have no monopoly of the meaning of Chinese-ness. Mrs Du’s lament only brings to the fore the boundaries within boundaries. These divide Han from Uyghurs (outside) but also separate Han who identify with China as a multi-ethnic state and those who understand themselves exclusively through Han-ness (inside). Even Mrs Du sharply distinguished Xinjiang from neidi by saying “in neidi you can see traditional Han customs preserved and celebrated but it is different here”. She was concerned at the loss of minzu-ness amongst Han in Xinjiang where “Xinjiang person” should not supplant but supplement Han-ness.

6.1.2: Religion and Food as Boundaries

The final chapter will discuss how Uyghurs are interpellated by Han as dangerous outsiders through the category of foreign-ness frequently explained through Islam. However, it is important to first consider how the pork taboo serves as a visceral everyday boundary between Han and Uyghurs but is imbued with identity politics to frame who is and is not a “real Chinese”. Smith Finley (2002) showed how Islam and the pork taboo are self-ascribed vessels of difference employed by Uyghurs to maintain boundaries with Han in Xinjiang. However, if we explore this boundary from the Other side, the Han ascription that Chinese people are pork- eating atheists is mutually constituted against Uyghur Islamic practices. Mao Zedong’s commitment to eradicating Han chauvinism was questionable given that he announced pork was a “” and induced Hui farmers to rear pigs (Gladney, 1996, p.26). Despite being a middle-income country, China regularly ranks in the top ten in the world for per capita pork consumption. The pig is often seen as intimately tied to land and family through the discourse of Han as an agricultural and peaceful people, as opposed to nomadic and dangerous. Eating or not eating pork may have little to do with ethnic relations but the meaning of this choice was consistently contrasted against the choices of the Other in interviews with Han. Mr Yi, a 40 year old male getihu explained the violence of July 2009 by saying:

It’s religion, all Muslims, Islam is all about killing people…ever since liberation - we liberated Xinjiang - we have had this problem. So now we hate them (points towards Erdaoqiao/Döngköwrük). It’s not possible to have unity and it’s not possible to have harmony. It’s just the government’s words. Han and Uyghurs can never have the same perspective, never have the same heart. (Points to slogan “unity is prosperity, separation is disaster”) It’s just a stupid government slogan. Do you eat pork?

Islam is self-selected by Han as a marker of difference and framed as a problem. Not all Han used the phrase “The Three Evils”. However, many demarcated community boundaries in ways which were informed by its association of Islam with outside-ness and as we shall see in the

200 final chapter, with violence. The pork taboo creates physical distance given that Uyghurs are forbidden to enter Han Chinese-owned restaurants, a central location of socialising and drinking for Han. However, the visceral act of eating pork is imbued with discourses of identity politics from below. These discourses demarcate the boundaries of China where commitment to its consumption of pork is a marker of national identity. During fieldwork, I was asked “do you eat pork?” countless times by Han interviewees and friends, who then expressed pleasure and even camaraderie at my affirmative answer. This then led to shaking hands, exchanging phone numbers, or being welcomed to stay in Xinjiang. Mr Qi, a late-20s getihu from Ürümchi, who loved listening to Russian disco music, told me:

Turban-heads is the name for Uyghurs, you know? The problem here is minzu division and minzu contradictions. Islam is just too extreme a religion. It’s bad, makes people use violence. I mean they don’t even eat pork, so they won’t eat in our restaurants. Do you eat pork? (happy at my response, says “good” and continues…) Pork is the best meat, everyone should eat it. They think it’s because it’s dirty but we say it is because their ancestors are pigs and they worship them.

The dehumanising notion that Uyghurs descend from a different species let alone race is often heard across China. “Turban-head” was the official Chinese designation for Uyghurs until the Sheng Shicai took over Xinjiang during the Republican period (Millward, 2007, p.208). Its persistence is not sponsored by the party-state today but it remains in use to exclude Uyghurs from the Chinese community. I had to explain that Uyghurs did not worship pigs to numerous students when teaching English in Shandong and to several friends in Beijing, even those studying at Masters level. In contemporary Xinjiang, Han self-employ these stereotypes as markers to demarcate racial difference from Uyghurs. The Islamic pork taboo is used to exclude Uyghurs from China but also from human-ness as “everyone should eat it”. Han-ness is thus objectivised as Chinese-ness and a common-sense or normal human mode of behaviour. Uyghur-ness by contrast is articulated as a form of violent extremism challenging the human- ness and peacefulness of China and pork. Miss Lan, the university teacher exemplified the objectivisation of Han-ness in contrast to the explicit ethno-centrism of the getihu. She explained that throughout history all Chinese people have defined the concept of family and home (jia) through owning land, a house, and pigs. In conjunction with standardised textbooks used all across China134, Mandarin language lessons at Xinjiang University in 2009 explained the Chinese character for ‘home’ and ‘family’ (jia; 家) as a “herd of pigs under a roof”. The lesson passed without mention of the pork taboo. This conceptualisation of family and home excludes Muslim inhabitants of Xinjiang from what is the nominally multi-ethnic culture of China.

Food in China is ordinarily a soft boundary in the sense that group members can share in other groups’ food without being converted (Duara, 2009, p.112). However, this boundary is

134 Boya Chinese (Boya Hanyu).

201 hardened and groups are constituted when these practices function as markers and are understood as constitutive principles of community (Duara, 2009, p.112-113). Christine Cesaro (2007) argued that Uyghurs assert their distinctiveness and superiority to Han Chinese by frequently comparing food preparation and styles to ‘Han’ methods. Uyghurs are engaging in a counter-hegemonic discourse to the “Han superiority implicit in the Han ‘civilising project’” (Cesaro, 2007, p.196). However, Han are engaging in these practices of boundary-demarcation to re-assert ethnic hegemony over Uyghurs. The violence of July 2009 and their representations may have shaped how this taboo is employed and conceptualised. During a year in Ürümchi 2007, it was common to be welcomed to Xinjiang by Han encouraging me to “taste the Xinjiang flavour” meaning Uyghur food of pilaf, laghman, nan, and kebabs. However, many getihu in 2009 warned me against this with one 40 year old male bendi ren explaining “Han don’t eat minzu food since 7-5, tuanjie is impossible”. Mr Wang, a late 30s male from Sichuan who had spent 11 years in Karamay as an oil executive said “even before 7-5, I really didn’t like Ürümchi. I don’t like Uyghurs. I really don’t like them. So now we don’t eat their food and won’t buy their things”. It was easily observable that many Uyghur restaurants had closed down in Han areas of the city and Han customers were staying away from those which were open. One interview with a local businessman based in the city centre explained that all Uyghur restaurants in the People’s Park (gongyuan beijie) area had been attacked on July 7th. They could only do business again when all Uyghur staff in contact with customers had been replaced with Hui135. Islam is being selected as a self-ascribed boundary between Han and Uyghurs but Hui Muslims are not subject to the same exclusions. Islam for Han is a representation which serves as a vessel. This enables people to “think themselves into difference” and maintain the concealed contingent meaning of boundaries which appear timeless and irrefutable (Cohen, 1985, p.86). Consumption of pork becomes a symbolic social practice to produce ethnic boundaries as an alternative form of nation-building.

6.1.3: Making Friends, Making Tuanjie

If we take willingness to make friends with other ethnic groups as a proxy for ethnic or minzu tuanjie, there is little evidence of any such feelings amongst Han. For example, Mr Qiang’s commitment to minzu tuanjie was purely an instrumental means to maintain the territorial integrity and international power of China. He framed himself as an ethnic Han and demarcated his community along ethno-linguistic lines. He had no Uyghur friends and clearly stated he did not wish to have any. His face-to-face community was entirely Han and did not correspond to the imagined community of inter-ethnic harmony offered by the party-state. One male getihu, aged 50, explained he was a local (bendi ren) but his laojia was in Hunan. When I

135 This practice was still observable by July 2012. I visited three restaurants on this street, all of which had Uyghur chefs but all-Hui service staff.

202 asked where was from, he replied “I am from Xinjiang; Chinese Xinjiang Hanzu person”. He explained he refused to have “any contact with Uyghurs” and would only speak to shaoshu minzu “who want to learn more Chinese”, which he said excluded Uyghurs.

Mr Yi, the getihu who blamed Xinjiang’s problems on Islam was keen to explain on our first meeting that he and other Han refused to make friends with Uyghurs. Mr Xin, a 40 year old bendi ren taxi driver explained that “tuanjie is impossible” and the violence in Xinjiang: “minzu contradictions, a minzu problem. Uyghurs and Han don’t get along…we Han drivers are not even willing to pick up Uyghurs136. Too much trouble. Tuanjie is impossible”. Even Mrs Du, who could speak Uyghur and Miss Lan admitted to me they had Uyghur colleagues but no real friends. Three young Han female residents studying Uyghur, all of whom had no Uyghur friends, claimed it was impossible due to “cultural difference”. Professor Du claimed that “how is your lamb?” was a common greeting amongst Uyghurs. When asked about this, several Uyghur interviewees laughed, saying Han know nothing about Uyghurs. Aynür, a Mandarin and Uyghur languages teacher, despaired, saying “I can’t believe people still have these attitudes”. This absence of social knowledge even amongst the educated indicated the social distance between two ethnicised and visceralised communities. However, it also illustrated the gap between the imagined multi-ethnic community offered by the party-state and the popular meanings imbued in daily practices by Ürümchi residents whose social interaction is largely intra-ethnic.

Minzu tuanjie amongst intellectuals appears to be a superficial form of multiculturalism involving a vocal commitment to diversity with little relevance to daily practices. However, the party- state’s conceptualisation of minzu tuanjie, discussed in chapter 5, is theorised through the objectivisation of Han-ness. Han cultural practices are taken as symbolic of the nation and non- Han cultures represented as peripheral. Mrs Du and Miss Lan’s commitment to ethnic unity should be thought of as superficial but their ideological commitment to minzu tuanjie as a way to convert Uyghurs and Han has been intimately shaped by party-state discourse. They saw eating mutton and learning basic Uyghur alongside fluency in Mandarin and interacting primarily with Han as a way of being a Xinjiang person. This, like minzu tuanjie, included Uyghur-ness as peripheral and Han-ness at the centre. On the other hand, all getihu interviewed entirely dismissed the notion of minzu tuanjie. They saw it as incompatible with the social practices of their daily lives, where social contact was largely intra-ethnic and through which they understood their mono-ethnic nation.

136 Mr Xin explained that I would have problems hailing a taxi in the city because people would mistake me for a Uyghur. This was occasionally a problem. It was invariably a problem when hailing a taxi with Uyghur friends. It was even an issue with my partner from Scotland because she often wore a flowery skirt, which people claimed made her look like a Uyghur. One driver who I stopped by stepping in front of his car when he was deliberately ignoring us said “I nearly didn’t stop. I thought you were a Uyghur but it’s ok, you are a foreigner (laowai)”.

203

6.2: (Re)Performing Uyghur-ness, (Re)Performing Inclusion / Exclusion

There is much diversity amongst people who call themselves Uyghur, particularly with regards to attitudes towards class, gender, and oasis-origin. However, near universal Othering of the Han and China was a boundary which informants chose to discuss on a daily basis and through which all politics were discussed. A synthesis of the top-down137 and bottom-up138 approaches to identity formation discussed in chapter 1 helps explore the interaction between official and unofficial identity politics and between competing visions of nationhood in Xinjiang. There is a growing interest in exploring the intersections between ascription from above and self- identification in Xinjiang139. Minzu are in some senses positionalities in which “minoritisation” may take place but it then takes a life of its own (Schein, 2000, p.96-99). It is important to explore how the balance between top-down impositions and bottom-up agency is contingent upon a number of factors including time period and social sphere. Some spheres of Uyghur social life are encouraged by the state, particularly those officially deemed “safe” such as traditional pre-Islamic religious practices, which make them more amenable to bottom-up analyses (Dawut, 2007, p.149; Bellér-Hann, 2001, p.9). Other areas such as the origins of Uyghur national identity, including researching pre-1949 debates on Uyghur-ness, are heavily circumscribed (Brophy, 2011, p.18). This restricts the ability to openly articulate Uyghur-ness from the bottom-up and creates challenges in doing research. However, the methodological choice made here was to build and adapt methods which reflect our ontological commitments. This chapter and the next thus utilises interviews with members of the postulated nation regarding their Self/Other-identification and its relationship with official discourse. The analysis will be divided into subsections which focus on how self-identification is articulated through spatial divisions, the city of Kashgar, and language.

Like the approach to discourses on Han identity in Ürümchi in the previous section, the identity politics of Uyghurs are best understood through Chow’s (2001) “semantic hybridity”. This follows Judith Butler’s (1994) approach that discourse builds on and challenges existing discourses from within rather than rejecting it outright. For example, the etymology of the ethnonym Uyghur is frequently and mistakenly taken by Uyghur nationalists as “tending to unite” (Brophy, 2011, p.1). The party-state has endorsed this etymology but claims it shows evidence of the Uyghur desire to be integrated into the Chinese nation through minzu tuanjie. The same discourse can be used in competing ways to contest nationhood. The question here then is not how the Uyghur came to be but how it continues to come to be through this

137 See: Gladney (1990; 1996; 2004), Rudelson (1997), and Schein (2000). 138 See: Smith Finley (2000; 2002), Newby (2007); Bellér -Hann (2008), Dautcher (2009), and Roberts (2009). 139 See: Bovingdon (2002; 2004; 2010) and Brophy (2011).

204 contestation. Uyghur-ness is produced through daily practices mediated by contestations over the boundaries of identity in ways which blur and merge the inside and outside of community. This approach follows Rogers Brubaker (1998)’s conceptualisation of group-ness as an event, which can form and disintegrate under different social conditions. People frame and practice their self-identifications in multiple and contextually contingent ways. The argument here is that nation-building and the boundaries which it entails in Xinjiang are the social conditions which are ethnicising and intensifying group-ness as both Han and Uyghurs resist conversion. Where the Han are responding to identity discourses which place them at the centre of the nation, Uyghurs are responding to those which position them on the periphery.

Indigenous sources written in the East Turki dialect from Xinjiang 1880-1949 used the ethnonym Turk almost universally (Bellér-Hann, 2008, p.27-29, 51). Uyghur historians have responded to official histories of Xinjiang as an indivisible component of the 5,000 year nation by re-asserting the ancient Turkic origins of Uyghurs through these types of historical sources (Bovingdon, 2004b, p.355). As chapter 2 showed, the writing of National history is a contested and contingent process of imagining boundaries but which secures the unity of a national subject through time (Duara, 1995, p.4). The claim that Uyghurs are “not a Turkic minzu”, in The 50 Whys for example (Ministry of Information, 2009, p.57), is a response to these self-identifications from below. The party-state seeks to delink the minzu category of Uyghur from the broader category of Turk. However, this category remains a powerful symbol amongst Uyghurs to contrast themselves as a civilisation against the category of Chinese. These claims elicit a response to the official where the meaning of Turkic is reconfigured to mean a timeless nation and a civilisation-state to offer an alternative to the timeless Chinese civilisation- state. The use of “Turkic” amongst Uyghurs today is not restricted to academic debates amongst historians. Many students both male and female from Ürümchi, Turpan, Kashgar, and Korla all told me “We are Turks, we are not Chinese”. One middle aged, female teacher explicitly referred to the 50 Whys to explain “minzu tuanjie is just party nonsense. We are Turks”. Interviewees from all social classes, from getihu to intellectuals, expressed some identification with being Turkic even if it wasn’t their sole form of self-understanding. These answers were in response to questions from a foreign researcher. The answer given to another Uyghur or Chinese person would probably refer to oasis of origin. These different self- identifications are not contradictions but indicative of the intimately contextual nature of self- identification and the Othering which it entails. The popular citation of Turk as an identity category responds to the discourses of China and Hanzu as a 5,000 year old, unbroken civilisation.

6.2.1: Discourses of Space and Identity Amongst Uyghurs in Ürümchi

205

Clifford Geertz theorised the “doctrine of the exemplary centre” to explain the power of the state (Geertz, 1980, p.13). Mass centralised rituals and symbols of the state are material embodiments and the basis of the process of legitimating a centralised political order. However, by calling themselves Turks, Uyghurs theorise spatial identities not through conceptual, national belonging but through the social bonds of language and religion understood through viscerality. Many Uyghurs express concern about visiting Beijing, the nominal centre of China, because they worry about the absence of visceral community. One female Masters student from Turpan said she could not go because “there will be no halal food”. A 30 year old male teacher from Hotan insisted we couldn’t even go north of the city centre of Ürümchi because “I don’t think there are any Uyghur restaurants in those areas”. There were other practical concerns regarding visiting the Eastern frontier. These included “the hotels don’t accept Uyghurs140” from a 40 year old Qumul-lik141 male scholar and “I won’t be able to make friends because of the cultural differences” from several male and female undergraduate students from Korla, Turpan, and Ürümchi. Mahigül, a student in her early 20s was a born and bred Ürümchi-lik from a professional family. She and her family were proud to be Uyghurs but also proud to be urban, modern Ürümchi-lik-ler. Her attitudes towards neidi exemplified the unfamiliarity and outside- ness that most Uyghurs expressed: “Inner China, Beijing, Shanghai, any other part of China, these are all the same to me. They are all Chinese- same food, same buildings, same life”. This understands the capital of the country and the objectivised symbols of minzu tuanjie and the Chinese nation such as the Forbidden City and the Great Wall as culturally unfamiliar and hostile social environments. Beijing as the centre of Chinese-ness is then a frontier on the periphery of how Uyghurs map identity and place. Uyghurs place their centre of understanding not in China but in Xinjiang. For Xinjiang to be taken as the “new frontier” for Uyghurs would mean re- mapping their self-identifications such that their centre becomes a periphery. Rather than building on existing discourses of identity politics amongst Han and Uyghurs, the nation-building of the party-state is directed towards revolutionary cultural rupture and a radical re-centring of self-identifications. It has been unsuccessful at supplanting the ethnic identity of Han who already accept the national category of China. It thus seems doomed to failure for a group who by rejecting the centrality of Beijing, reject the party-state’s conceptual community of Zhonghua Minzu with the Han at the centre.

In Ürümchi, Uyghurs conceptualise the space they inhabit in different ways from both the party- state and the Han residents interviewed. When Han frame the region as a frontier to be liberated and civilised, they understand it as Chinese but on its periphery with inner China as the centre. Amongst Uyghurs there is no reference to frontiers when referring to Xinjiang, only

140 This claim is also backed by evidence in adverts and municipal government regulations for certain hotels as Rayila (2011) has shown. 141 The suffix –lik is added to place names to indicate a person from that place in the Uyghur language. A Qumul-lik is then a person from Qumul where Qumul-lik-ler is the plural.

206 centrality. These include references to their homeland (weten) meaning Xinjiang. Turghun Almas’ book Uyghurlar (Uyghurs) famously framed the history of Xinjiang and Uyghurs as Central Asian. This instigated a week long conference to condemn the book followed by public burnings (Bovingdon, 2010, p.97). When discussing Uyghurlar and its framing of Uyghurs as Central Asian, a young businessman, Mukhtar laughed saying “this is Central Asia. Han only came here recently”. Mukhtar, in his early 20s, had recently graduated and turned his hand to business. He represents an amalgamation of Joanne Smith Finley’s ideal types (2000) of the “politicised teenager” and the “young urban male intellectual”. He listened to foreign radio to get the “real news”, frequently criticised the government, and constantly worried about the future of Xinjiang. He also read academic literature on Xinjiang and hoped to study at doctorate level. For Mukhtar, Islam and the Turkic languages of Central Asia are not peripheral but central. Mukhtar was challenging the peripheralisation of Xinjiang and its timeless Chinese-ness. He was highlighting how what is a “new frontier” for Han is an ancient centre for Uyghurs. This was a (re)performance of Xinjiang as timelessly Chinese where the text from above was turned against itself so that Xinjiang became timelessly Central Asian. These competing perspectives produce dynamic contestation over the meaning of Xinjiang but which remain concealed if the category is left as a taken-for-granted unit of analysis. When I asked Mukhtar and Mahigül if they were Junggo-lik (“person from China”), they found this amusing. Mahigül would say “I am a Uyghur in China”, particularly to new foreign friends, explaining to me that she didn’t want them to think Uyghurs were Han or Chinese. The refusal to be Chinese echoes the ambivalent position of Xinjiang in the Chinese empire. Since the Qing dynasty it most certainly has been in China but it is not yet of it from Uyghur or Han perspectives. The party-state’s interpellated categories of identification, including that of Xinjiang person, appear to be failing to hold any broad resonance with the population of Xinjiang.

6.2.2: Kashgar as the Centre of Uyghur-ness

Every Uyghur interviewed from all classes selected Kashgar and or their hometown in response to the question “what places are important to Uyghurs?”. When followed up by asking “what about Beijing?” most assumed this was a joke while one asked if I was picking a fight. Kashgar, as the home of the Id-Kah Mosque, is generally considered to be the centre of Uyghur Islam and frequently mobilised as a symbol of Uyghur-ness. One of the most popular folk singers in Xinjiang today, Perride Mamut, sings “she welcomed us so warmly, oh beautiful Kashgar” in the song “Coming to Kashgar”.142 The CD and DVD markets across Xinjiang have spawned many ‘independent’ artists, such as Nurmehemmet Tursun, Perride Mamut, and Abdurehim Heyit, whose recordings are very popular and sung solely in the Uyghur language. They play their own music but sing a mixture of their own lyrics and traditional folk poems, often in Chagatai. They

142 The translator of these lyrics as well as those of Abdurehim Heyit requested to remain anonymous.

207 blur the assumed lines between pop as modern and forward looking and folk as traditional and representing continuity with the past. They are popular because they are folk artists. In many ways the official Chinese sphere and unofficial Uyghur sphere appear as “parallel worlds” through music (Harris, 2005). The ‘Coming to Kashgar’ VCD video shows Perride delighted to leave Ürümchi by modern means of a plane. She flies back home to Kashgar to walk among the back streets of the old town and return to her community and her family. This (re)performance of Uyghur hospitality inverts the text of Chinese modernisation discourse. It positions Uyghur tradition as superior and desirable instead of backward and in need of transformation. However, Uyghur-ness is being located outside of the capitals, both Ürümchi and Beijing.

Perride’s “Coming to Kashgar” is all the more powerful when we consider the demolition of Kashgar old town, which was taking place in 2009. The demolition is explained by the party- state through the “modernisation” of Xinjiang. However, the demolition is also represented as the means to “rescue” Uyghur culture as “the housing style after ‘renovation’ will preserve the original housing style and the cultural characteristics” (State Council, 2009b, p.29-30). Perride’s video is a “taking up of the tools where they lie, where the very ‘taking up’ is enabled by the tool lying there” (Butler, 1999, p.185). These tools are simultaneously taken from performances of Uyghur and Chinese identity such that Uyghurs are represented as traditional but in a way that is superior to Chinese modernity and which excludes Han from the imagined centre of Uyghur-ness. Ambiguity in political and cultural expression is always safer in China. Since the crackdown at Tiananmen in 1989, idealism and overt resistance has given way to negotiation and “de-politicisation” for political purposes (Baranovitch, 2003, p.36-38 & 44-45). However, in Perride’s song and in everyday speech, the party-state’s vision of modernisation and nation- building are being turned against themselves as old Kashgar is mobilised as a symbol of Uyghur-ness outside Chinese-ness. The mobilisation of Kashgar as symbol can be read as a romantic vision of Uyghur-ness from outside the city but performed in the city. This is less a rejection of modernity per se and more a commentary on the model of modernisation offered by the Chinese party-state where Uyghur-ness is excluded from modernity and from the city. However, in Perride’s video, those features deemed traditional, rural, and ethnic are being placed inside the city and inside an alternative vision of ‘Uyghur’ modernity.

6.2.3: Ana Til: Language and Uyghur Self-identification

Every Uyghur interviewed during this research identified Uyghur language as central feature of “Uyghur heritage”, including Mukhtar, Mahigül, and Aynür. Most expressed serious concern regarding its elimination as a medium of instruction from the education system. One important song, which was cited by several interviewees as both a great song and an important message, was Ana Til (“Mother-tongue”) by Abdurehim Heyit. Most Uyghurs and every musician in

208

Xinjiang know the deep resonant singing style of Abdurehim Heyit and call him a “Uyghur Dutar143 Master”. Almost every VCD shop I visited recommended him first as both the best and most popular Uyghur music. The seriousness in his lyrics, facial expressions on his album covers, and his mournful singing, resonate with how the Yi artist Lolo used his album covers to express “sternness” as a rejection of the representation of all minorities as “happy, smiling natives” (Baranovitch, 2003, 85-86). Abdurehim made a name for himself through his lyrics but also by operating outside official channels, giving away free cassettes of his recordings in the 1990s. This was before the introduction of laws in 2001 requiring all artists in Xinjiang to submit all lyrics before performing or recording them which left Abdurehim Heyit unable to tour or release recordings (Bovingdon, 2010, p.96-97). Abdurehim’s songs included “Stubborn Guest”, a tale of an old man pleading for a guest to leave after over-staying his welcome. This “guest” is a commonly used allegory for the Han Chinese presence in Xinjiang in contemporary Uyghur music and literature (Bovingdon, 2010, p.98-99). The song, Ana Til, specifically celebrates Uyghur-ness through the symbol of the Uyghur language, imbued with imagined viscerality against the Chinese discourse of modernisation and the policy of “bilingual” education:

I want to respect the man, who knows his mother-tongue, I want to give gold to the man, who knows his mother-tongue, If this mother-tongue is in America or Africa, I would spend thousands of money to go there, Oh my mother-tongue you were given by our ancestors, I want to be proud of you in this world.

The song was a particularly popular one and Mukhtar was keen to stress to me “you know this is not about mother-tongue in general, it’s about saving Uyghur”. The song imagines a long and unbroken history of the Uyghur community through the transmission of the Uyghur language “by our ancestors”. This performs the boundaries of Uyghur-ness and community (“the motherland”) through the Uyghur language. It is performed outside the nation-building discourse of Zhonghua Minzu based on the symbols of modernisation and Mandarin Chinese. This prioritisation of language in the symbolic construction of community reflects a broader anxiety about the boundary-transgressing practices of modernisation discourse and “bilingual education” policies. Aynür, an early 40s teacher of Uyghur and Mandarin spent most of her adult life in Ürümchi. Fluent in Uyghur and Chinese, she took self-conscious pride at being “between cultures”. She represents the individualist. Aynür expressed a strong sense of ethnic identity but our conversations revolved around how community, ethnicity, and politics affected her life. As an employee of the state, she is a useful counterpart to Mukhtar. “Language is a way of thinking” said Aynür before stressing that Uyghurs who could not read their mother-

143 The Dutar is a two-string lute played in Xinjiang and which variants of are seen across Central Asia.

209 tongue had “no identity”. Abdullah, a 40 year old self-employed businessman explained “Language is about thought and those who are learning two languages like they do here are finding themselves confused, not able to express themselves or understand themselves properly, there is a contradiction between their thinking-religion and their language”. These are regularly heard claims of language ‘infection’ amongst Uyghurs. They understand the Otherness of China as a source of infection to Uyghur-ness much as Uyghurs are represented with regards to China. Mukhtar asked “how can you know your culture if you can’t read the books of your elders? It’s a national identity, it’s just a shame we are losing it”. Community here is predicated on a common mother-tongue and it is (re)performed as outside Zhonghua Minzu built on Mandarin.

Mahigül, like all interviewees, used Uyghur language with all her friends and family. She refused to make friends with anyone in Ürümchi (other than foreigners) who could not speak Uyghur or would not at least attempt to do so. She ethnicised language use and, like Mukhtar, employed an elision which equated the ethnic category of Han with the national category of Chinese: “I only speak Chinese to Chinese, Han Chinese to Han Chinese. I speak English to foreigners, I speak Uyghur to Uyghurs”. When I asked Mahigül why she felt language was so important to her identity she defined herself and Uyghur-ness solely in ethno-linguistic terms:

We are Uyghur, we speak Uyghur. We are not Uyghur speaking Chinese. Uyghur is ours. Uyghur language also has many centuries of history so it deserves to be kept. It’s part of our culture. I mean if we don’t speak Uyghur in the future, we can’t call ourselves Uyghur. We’d be Chinese.

Mahigül often articulated Uyghur-ness through modernity by saying for Uyghur culture to survive it would have to “create new things” and “forget about the Muqam”. However, like Mukhtar, she was (re)performing Uyghur-ness as nationhood through the discourse of a long and unbroken linguistic history to be preserved. This ethno-linguistic demarcation of the boundaries of community defines inside-ness in terms of mastery of mother-tongue (Uyghur). Through the elision of the categories of Chinese and Han, the outside was China framed as a Han nation-state. To lose one’s mother-tongue would mean the internal/external boundary is transgressed and one simply becomes a member of another community- Zhonghua Minzu. This does not perform minority-ness or an ambivalent inclusion. This performed a boundary between Uyghur-ness and Zhonghua Minzu-ness such that inclusion is equated with Sinicisation meaning Hanicisation. This is an example of how we cannot understand performances of Uyghur-ness solely through the dichotomising categories of Han and Shaoshu Minzu. The party-state’s discourse on the imagined community of Zhonghua Minzu had failed to convince Mahigül that she shared a community with non-Uyghurs. This was because it failed to correspond with her daily practices of language use. Mahigül used Uyghur for communication and making socially connections but used Mandarin Chinese to participate in necessary but anonymous economic transactions. She explained the need to speak Mandarin by saying “it’s the basic thing to live in

210

China…we have to buy things, we can’t always buy everything from Uyghurs”. She contrasted these interactions and relationally imbued them with symbolic meaning. Uyghur-ness was imagined as a form of viscerality with the constant capacity for social connection where China was an unavoidable but negotiable background.

As discussed in previous chapters, Ma Rong links China’s modernisation and international strength to a shared identity and the need for minorities to learn Mandarin Chinese. Ma Rong argues that the current policy of bilingual education, with Mandarin as the sole language of instruction, is essential for China to “have the ability to develop into a modernised nation” (Ma, 2009, p.240-241). Ma Rong (2007) argues that it is culture and not ethnicity which have historically defined social distinction in China. The distinction between “civilisation” and “barbarians” in ancient China and different minzu in contemporary China then orders the nation (Ma, 2007, p.7). The barbarian/civilisation distinction is not understood as a distinction between different civilisations but between “highly developed and less developed ‘civilizations’ with similar roots but at different stages of advancement” (Ma, 2007, p.5). This highly influential framework positions identification with China as positively correlated with economic development such that ancient “barbarians” naturally became Chinese by being developed and thus, learning or assimilating into Chinese culture (jiaohua). This framework orders the Chinese nation as bounded into groups who are more and less Chinese and more and less modern as opposed to different. Zhonghua Minzu, thus understood, is based on Louisa Schein’s (2000) internal orientalism. This model of nation-building requires the inclusion of ethnic minorities as behind the modern, Han centre, which offers prosperity.

In 2008, Ma Rong presented his work on “bilingual education” at Xinjiang Normal University. According to one attendee, Ma was heavily scolded by local Uyghur Professors. Ma’s models (2009) based on gradual introduction of Mandarin Chinese according to different regional needs and resources were said to be a “fabrication”. The programme in practice aims for rapid overhaul of the education system and has created considerable social tension. In 2011, 20 teachers at a college in Ürümchi refused new lower-rank jobs due to lack of Mandarin Chinese ability, with one stating “we are Uyghur, we should keep our language for the preservation of our culture” (RFA, 2011). This counter-securitisation of identity frames those policies which aim to increase security as threats to the ongoing existence of self-identification. Uyghurs are securitising their understanding of ethnic boundaries in response to the exclusions of the defining symbols Uyghur-ness by the party-state.

Chinese nation-building does not acknowledge difference on a qualitative level as such; minorities are merely quantitatively behind the Han but can be developed and assimilated (Barabantseva, 2009). This hierarchical, pluralist model of nation-building marks different

211 ethno-linguistic communities as more and less modern. These articulations of levels of modernity linked to ethnic identities seek to rupture and convert how people identify themselves. Mukhtar, like Mahigül, Aynür, and Abdullah defined community as ethno-linguistic and being Uyghur was about being able to communicate, read, and write using the Uyghur language: “Uyghur people look so different, some of us look like Europeans, some of us look like Chinese, some of us look like they are from . The only thing that keeps us together is our language…I’ve heard there are, even Christian Uyghurs144 which I’ve heard there are many, if they think they are Uyghurs and they speak the same language, and share the same culture as me then yeah they are Uyghurs, they’re my friends, they’re my people”. This excluded non- Uyghurs and those Uyghurs who could not speak Uyghur such that Mandarin Chinese and the ‘nation’ were relationally framed as the economic, the imposed, and the outside against the social and the inside-ness of Uyghur-ness. Mukhtar said the Bilingual Education system was a “total failure” because it was not bilingual: “we are losing our language…they want to kill our language slowly then assimilate us”. To understand the “wholeness” of the text of nation- building requires we explore elites and subaltern as mutually embedded in constitutive relations. Mukhtar was positioning Uyghur-ness outside Chinese-ness. However, the “wholeness” of Mukhtar’s speech could only be understood through the tools he was taking up to turn hegemonic discourses against themselves and define an alternative vision of nationhood. Nation-building was being rejected with the tools through which it sought to define the nation with. Language, civilisation, and nation were (re)performed as synonymous but evidencing the Turkic civilisation of Uyghurs outside Zhonghua Minzu.

Mukhtar would say that cultures disappear as part of “nature” but that “assimilation” in Xinjiang was “not natural”, it was politically imposed. He objected to assimilation because he did not want to form a community with the Han. He would say “these things happen but just not them”, listing personal, visceral experiences before relating these to community. He expressed disgust at his younger cousin’s action of leaning over his parents to reach for their schoolbag. A “real” Uyghur apparently would never lean into the physical space of their elders. This was attributed to studying in Chinese. His cousin had seemingly “forgotten his culture” and this was the future for Xinjiang: “they are supposed to have Confucius who tells them to respect their elders but they just don’t”. “Bad individual traits” of Uyghurs educated in Mandarin-medium schools are often framed by other Uyghurs in Xinjiang as evidence of and the acquisition of “Han disposition” 145 (Smith Finley, 2007, p.224 & 229). Mukhtar, Aynür, and Abdullah all used Chinese speaking ethnic minorities such as Mongolians, Manchu, and Xibe as examples of the future for Uyghurs. They both claimed that these ethnic groups “disappeared”

144 This particular comment is not representative of elder generations of Uyghurs, who place the role of religion more centrally in their self-identification. However, it should be noted that Mukhtar still considered himself a Muslim. 145 Chapter 7 will explore this theme more fully.

212 with the “disappearance” of their language. In reference to a New York Times article Mukhtar said he did not fear “Xinjiang becoming another Tibet” but was concerned it “would become another Manchuria”. By this he meant the civilisational boundary between Hua and Yi had disappeared with the decline of the Manchurian language. Abdullah explained identity similarly, saying that the Xibe were “like other minorities - very assimilated, very Hanicised (Hanhua). It’s because they don’t have their own language anymore, they disappear”. Aynür in reference to north-eastern ethnic groups, particularly Manchu, Mongolians, and Xibe concurred: “many of these minzu have become extinct throughout history because their language died out. They are very much like Han”. The ancient binary division Elliot (2011) and Leibold (2007) discuss between Hua and Yi is reiterated and performed in official party-state discourse, which holds little appeal for the “barbarians” who are now to be converted. Uyghurs are (re)performing this binarism as a defining civilisational boundary which divides Turk from Chinese. They do so in ways which reject the centrality and superiority of Hua and seek the rejuvenation of an ancient and parallel Turkic civilisation. It is a boundary which defines civilisational belonging and thus Chinese nation-building is seen by Uyghurs as a transgression of Self and Other.

If we gauge popular sentiment through popular culture, protest, and ‘unofficial’ discourse we can say that nation-building is being turned against the party-state by Uyghurs who use it to reinforce cultural difference. Popular culture, such as the music of Abdurehim Heyit and Perride Memmut is performing Uyghur-ness in ways which challenge the discourses of modernisation and the policy of bilingual education. The party-state’s centring of Han-ness through minzu tuanjie is being (re)performed by Uyghurs as outside their centre. This centre is defined through Turkic-ness and Islam, the features specifically cited by the party-state as propaganda of “The Three Evils” discussed in chapter 3. Uyghurs in Ürümchi, like getihu Han, see the model of multi-ethnic community with the Han at the centre offered by the party-state as incompatible with the social practices of their daily lives. Social contact is largely intra-ethnic and daily life is imbued with ethnic meanings. Party-state performances of Chinese-ness are having the opposite effect of that intended. These performances are (re)performed as modes of resistance. Uyghur-ness, like Chinese-ness, is enacted through discourses of timelessness, civilisation, and ethnicity.

Conclusions

This chapter has shown how the party-state has been unable to assert a monopoly over the boundaries and meaning of identity in Xinjiang. Ascriptions from above and self-identifications from below in Xinjiang are not reducible to one another. Neither the official nor the unofficial can be placed in a position of causal primacy when discussing identity politics. Understanding the “wholeness” of the dynamics of identity contestation requires analysis of the relations

213 between the two. Instead identity is taken as a political discourse and identification is an ongoing process of negotiation and contestation from above and below. The incompatibility between official politics and how people identify themselves coupled with the inherent flexibility of language is by no means unique to China. However, what makes this case particularly problematic is that most official and popular discourses are alternative modes of essentialisation. These essentialisations attempt to articulate timeless discrete communities in the face of social contingency. The party-state does define who belongs to what minzu but its attempts to define what these categories mean are taken out of their hands by people who use this discourse for their own ends.

The first section showed how Han in Ürümchi were in many ways contesting their own identities by how they framed the position of Xinjiang in China. Intellectuals lean towards adopting the concept of minzu tuanjie and the objectivisation of Han-ness this entails to promote nation- building in the region. However, the working class and getihu Han tended to reject it altogether as contradictory to their nation defined through ethno-national terms. This was as much a contestation about the external boundary over whether Uyghurs ought to be included in community as it was about what it means to be Han and Chinese. The conversion of Uyghurs in minzu tuanjie implies China is Han-centred but inclusive of those who wish to be converted through mastery of Mandarin Chinese and self-identification as Zhongguo Ren. The ethnic identification of getihu on the other hand defines Han-ness primarily as lineage. This is a form of ethno-nationalism which superimposes pre-1949 cultural discourses onto post-1949 minzu. The explicit ethnocentrism of great Han chauvinism remains implicit in the Han-centred concept of minzu tuanjie. By this logic, Uyghurs cannot be converted for their lineage lies outside of China. Instead Uyghurs are a modern day equivalent of the barbarian Other which they defined themselves against. These ethnocentrisms exclude Uyghurs and shape the official account of the multi-ethnic nature of Chinese nationalism. There is very little academic discussion or public debate in China on this issue of what would be called “separatism” if it was not an identity held by the ethnic majority in today’s China. It is important to avoid essentialising Han as if their multiple and often divided voices are represented by the interpellations of the party-state. However, it is also important to consider the dangerous implications for security when one group (Uyghurs) which protests for greater cultural rights can be stigmatised through the discourse of “terrorism”. Particularly when another (Han) is not subject to securitisation but often taken for granted as a reservoir of patriotic energy and loyalty to the party.

The second section showed how Uyghurs turn 5,000 years of Chinese history against the party- state. Uyghurs imagine a timeless Turkic civilisation running parallel to Chinese civilisation, rather than above or below it. Party-state discourse did not invent the category of Turk but it has given it meaning by offering it as a means to resist being Chinese. Uyghurs in Ürümchi, like

214 getihu Han, see the model of multi-ethnic community with the Han at the centre as incompatible with the intra-ethnic social practices of their daily lives. Nation-building is being turned against the party-state by Uyghurs who use it to challenge the centring of Han-ness through minzu tuanjie. China is being (re)performed by Uyghurs as outside their centre, defined through Turkic-ness and Islam. Party-state performances of Chinese-ness are having the opposite effect of that intended as they are (re)performed such that Uyghur-ness and Han-ness are enacted through discourses of timelessness, civilisation, and culture. Popular Han and Uyghur framings of community are both drawing on language, lineage, and culture as vessels to demarcate boundaries between the two. Minzu tuanjie transgresses this boundary on both sides where Han turn to lineage to distinguish themselves from Uyghurs who see themselves as Turkic.

Boundary contestation in Xinjiang may be about the intrinsic value of the boundary itself but it is also about how to conceptualise community. Han and Uyghurs are resisting the idea of their nations as “imagined communities” of people who have never met united by territorialised conceptions of nationhood. Instead, they seek to assert viscerality where language and lineage are framed as authentic and objective sources of self-identification. Nation-building in Xinjiang is failing to convince both the Han majority and minority groups that they are part of the same national community. They continue to Other each other and define their communities along ethno-linguistic lines (Han China vs. Turkic Central Asia). These ethnic categories should not be reified. However, the constant repetition of these themes by individuals who identify themselves as Han and Uyghur suggests that these are significant social trends with important social consequences. The party-state’s discourse of Chinese-ness as non-ethnic and inclusionary is unlikely to find supporters amongst Uyghurs because Han-ness remains an exclusionary category based on lineage and divides the “sedentary” Central Plains from the “nomadic” West. While minzu tuanjie speaks to Han nationalists by placing them at the centre, it fails to appeal to them because they seek the exclusion of Uyghurs from community, rather than their conversion.

Miss Lan, the Mandarin Chinese teacher explained that “minzu tuanjie is an ancient tradition of the Chinese people” but this can only speak to people who already understand themselves as Chinese. James Leibold’s (2010) three models of ethnic inclusion in China (state-sponsored Leninist minzu policies, Han ethnocentrism, and Confucian ecumenism) should not be seen as mutually exclusive social practices. The argument here has been that all three are present in nation-building and popular Han (re)performances of nationhood in Xinjiang. The drive to categorise minzu (Leninism), unite them as Chinese (Confucianism) but place the Han at the centre (ethnocentrism) are of equal importance in Xinjiang. Han are thus in a powerful position to select from an array of seemingly incompatible approaches to identity politics and exclude

215

Uyghurs in different ways. This chapter showed how ethnocentrism amongst Han and Uyghurs takes primacy in discourses of identity. The final chapter will explore how threat and security are framed from below in Ürümchi. It will discuss the contradictory framings of identity-security where Han see the “backwardness” of Uyghurs as a threat to China’s modernity while Uyghurs see “modernisation” as a threat to their tradition.

216

Chapter 7: (Re)Securitising Ethnicity, (Re)Performing the Nation

Introduction

The last chapter showed how Uyghurs and Han in Ürümchi are resisting Chinese nation-building through the (re)performance of official discourse. The nation-building model which places the Han at the centre of a multi-ethnic community performs boundaries through the exclusion of the “inside/outside Three Evils” as an existential threat to a sense of Chinese “we-ness”. Earlier chapters focused on the securitisation of Zhonghua Minzu performed by the party-state. The external boundaries of the Chinese state and the internal boundaries of ethnic hierarchy were officially framed as “beyond politics”. Security and identity are different ways of talking about the same thing. Security and the power relations concealed by the categories it employs produce the boundaries it claims to reflect. Exclusionary categorisations relationally produce security and nationhood through Judith Butlers “negative elaboration” and Thongchai Winichakul’s “constitutive outside” (Butler, 1999, p.132; Winichakul, 1994, p.167). However, these discourses are being turned against itself by Han and Uyghurs. These groups tend to frame Self/Other-identification through ethnic boundaries and often ethnocentrism. The analysis presented thus far has shown how daily practices are being imbued with ethnicised meanings to reinforce these boundaries. This chapter will explore how these different meanings are securitised by Han and Uyghurs where minzu tuanjie is framed as an assimilative transgression of ethnic boundaries. This securitisation from below performs the boundaries of Self and Other through discourses which centre on the threat to we-ness. This chapter is about (re)securitisation and looks at how politics from below speaks to official politics. Nationhood is (re)securitised when postulated members of the nation (re)perform identity-security discourses to articulate alternative political communities.

Understanding the “wholeness” requires an analysis of the relations between securitisations from above and below. “Wholeness” refers to the mutually constitutive nature of identity and security discourses within the dynamics of identity contestation. These contestations are often concealed and obscured if we equate ethnonyms and national categories with identities. This final chapter will discuss the (in)security dilemma which emerges as the party-state imposes its framings of identity-security on people who understand themselves in alternative ways. The different approaches to studying Xinjiang discussed in the literature review were largely framed through Bellér-Hann’s (2007) division of Central Asian historians and anthropologists on the one hand and Sinological political scientists on the other. These literatures tend to construct research questions on Xinjiang and Uyghurs pertaining to identity or security respectively.

217

However, the approach here is to treat identity and security as mutually constitutive performative discourses.

The chapter is divided into two sections, which focus on how Han and Uyghurs (re)perform the boundaries between Self and Other through the official discourses of identity-security in Xinjiang. The first section will explore how Han in Ürümchi redeploy the party-state’s discourses of danger to articulate their own conceptualisations of threat and security. This will be termed (re)securitisation. Han frequently use these discourses to exclude Uyghurs as an ethnic group as a threat to identity-security thus contradicting the CCP’s model of a multi-ethnic national community. This (re)performs China as an ethnic Han nation building on discourses of lineage and development discussed in previous chapters to resist multi-ethnic nation-building. The second section will analyse how Uyghurs in Ürümchi (re)perform official discourses and unofficial Han framings of what it means to be Uyghur through (re)securitisations from below. This explores how Uyghurs turn the discourse of nation-building against itself to invert threat and security where the party-state is framed through discourses of danger as a threat to Uyghur-ness. This chapter will explore how Han and Uyghurs conceptualise and identify threats to their person and their sense of Self in ways which are shaped by official discourse but which resist this model of nation-building. The argument is that the party-state, Han, and Uyghurs all perform different identities through competing discourses of danger. The boundaries through which social practice is being framed are being contested in terms of threat and security from above and below. (Re)securitisation and (re)performing nationhood are part of the same process of resistance to the party-state’s model of China, which exacerbates identity-security concerns amongst both groups who see themselves as marginalised.

7.1: Securitising Han-ness, (Re)Performing China

The analysis here will explore how discourses of identity politics which target Uyghur-ness as a threat shape and are shaped by securitisation from below. This section will explore how Han re- deploy the party-state’s discourses of danger to articulate threat and security in Ürümchi in ways which resist its model of nation-building. It is divided into three analytical sub-sections which focus on the securitisation of spatial divisions, the ethnicisation of violence, and ethnic nationalism. Ürümchi is spatially divided between the Han-populated north and Uyghur- populated south (Taynen, 2006; Smith, 2002). After the violence of July 2009 many Han who did live in the south moved north. An estate agent for the area explained that house prices on Jiefang Lu (Liberation Street), the thoroughfare linking Nanmen (South Gate) a busy retail area and Erdaoqiao/ Döngköwrük, had dropped by 50% almost overnight. Students and getihu

218 described the south of the city to me as the “enemy occupied district” (diqu)146. This word play on the real estate division between a, b, c and d districts, indicated how danger was being openly articulated from below through ethnicity. Friends and enemies, the inside and outside of community, were being articulated through an ethnic antagonism between Uyghurs as a threat and Han as security.

7.1.1: Securitising and Ethnicising Space Amongst Han in Ürümchi

“7-5” was a hot topic of conversation when I arrived in city in September 2009. For several months, almost every individual of Han ethnicity I met raised the issue with me without prompting. The open-ness of Han residents in discussing political topics reflected the asymmetrical power relations in the city. Ross Anthony (2011) showed how these relations placed Uyghurs under surveillance were not applied to the Han. These asymmetrical power relations restrict the space in which Uyghurs are willing to discuss issues pertaining to politics and ethnic relations (Anthony, 2011, p.51). Han residents were struggling with insecurity during this period as many still feared for their physical safety. However, they framed insecurity of their bodies through identity politics. One mid-20s Ürümchi-born and bred Han office worker had still not left her apartment block three months after the event. She warned me against visiting the Uyghur parts of town because it was “dangerous” and “they caused ‘7-5’”. At this time there was a presence of tens of thousands of PLA troops and the PAP on the streets of Ürümchi. Nevertheless, many interviewees said I was “fierce” (lihai) for walking the streets and facing the security threat they saw located in the Uyghur district Erdaoqiao/Döngköwrük.

On the day of the 60th anniversary of the founding of the PRC, the Mandarin language teacher, Mrs Xi, urged me not to go to the city centre. Instead, she recommended I stay at home to watch the TV transmissions from Beijing: “I urge you not to go. You can’t go to places with lots of people. 7-5 has made Han scared”. She claimed it was perfectly safe for minorities to go but not for Han thus framing security and ethnicity in inseparable terms. Ethnicisation of daily life and the boundaries of security and danger were found amongst most Han interviewed during this period. This was securitisation of an ethnic Other to the extent that they chose not to leave the “security” of their ethnically segregated living area to avoid contact with danger projected onto Uyghurs. In an official welcome to students at a university, one teacher even told her students that “all Uyghurs like fighting and carry knives”, so “don’t go to Erdaoqiao alone”. These exclusionary statements exemplify the open-ness which Han were permitted to express ethnicised framings of daily life in the city.

146 Residential districts were divided into a, b, c, and d. D-qu literally meant ‘district D’ but in spoken Mandarin this sounds like diqu (“enemy-district”).

219

Mr Li, a 43 year old local-born labourer, who despite his love of the military and action films, recounted his fear of violence from Uyghurs. Mr Li told me “Xinjiang is such a good place – even better than Beijing. It’s because the people are so warm-hearted. They want to help each other because they come from all over China. But don’t got to erdaoqiao, it’s very dangerous. Rich Uyghurs pay poor Uyghurs from the south to come here and riot”. When I asked Mrs Xi who was to blame for “security problems” she also said “rich people”. These (re)performances of danger were drawn from the party-state’s discourses on “the inside/outside Three Evils”. However, for Mr Li and Mrs Xi the inside/outside enemies were not simply organised terrorist groups but Uyghurs both poor and rich. One early 40s, male shop-owner warned me on our first meeting that “East Turkestan forces” were rising up to separate Erdaoqiao from the “motherland”. The insecurity caused by being witness to the mass violence of July was visceral and would be present with or without government commentary. However, the popular framings of the violence drew from the party-state’s construction of an ever-present threat in the form of the “Inside/outside Three Evils”. The party-state had claimed that “7-5” was “not a minzu problem, not a religious problem but a political problem of defending the unity of the nation” (Ministry of Information, 2009, p.4). However, threats were being (re)performed by Han from below as specifically ethnic (i.e. Uyghur). Uyghurs were being represented through (re)performances of “The Three Evils” as a threat to the body politic where boundaries and security are demarcated through ethno-nationalism.

7.1.2: Han Ethnicisations of Violence in Ürümchi

Performances of security and boundaries were ever-present as we saw through analysis of the official discourse on “7-5” and the militarisation of the city’s streets in Chapter 4. Violence by Uyghurs against Han was framed as an existential threat to be secured against where that of Han against Uyghurs was officially silenced and only spoken of in private. Some chose not to acknowledge the events of the 7th July, such as Mr Chen, a local-born 43 year old getihu, who said “they were all separatists - killing children, elderly, and ordinary people. It was very cruel. It was Uyghurs, all Uyghurs”. The middle class Han interviewed, such as Mrs Du, also avoided this topic. Most getihu, however, were more outspoken. A 45 year old getihu male from neidi laughed whilst telling me “the rioting, looting, and killings of 7-5 were very cruel but on 7-7 we Han got together to kill Uyghurs”147. Responses such as this were common amongst getihu Han over the course of this fieldwork. Mr Hu, the taxi driver and proud Xinjiang person, told me that Xinjiang was still not stable due to the “cruelty” of Uyghurs on July 5th. I asked about the killings on the 7th July but he laughed and said “no, there was no trouble at all on the 7th. It

147 Interviewees would explain how “we” killed Uyghurs but when asked if they were personally involved the interviewees chose to continue their narratives uninterrupted, change topic, or laughed and said they couldn’t tell me.

220 was fine”. I reframed the question, saying “some of my Han friends said there was a lot of violence on the 7th as well”. Mr Hu responded, “No, on the 7th us Han got together to kill Uyghurs. It was nothing, there was no problem”. The visceral experience of violence may appear inherently material and its security threat self-explanatory. However, its meaning was being mediated by discourses on nation-building. Violence against the national majority is framed as a security threat to the person and to the nation where that against minorities is “no problem”.

In the same way that the ‘doing’ of community matters less than the ‘thinking’ about what its ‘doing’ means (Cohen, 1985, p.98), security is best understood as social practice through which identities are articulated and contested. On July 5th 2009, local-born taxi driver, Mr Yan, had to rescue his daughter from Uyghur rioters. Rioters had tried to kill them both and his daughter was stabbed in the arm during their escape. Mr Yan explained how he conceptualised security by saying “we had to attack back. If they attack your family, you have to kill them. So on 7-7 we Han got together and attacked them”. “They” and “we” were framed entirely through the prism of ethnicity. “They”, an entire ethnic group can be essentialised as a security threat to be eliminated entirely through the visceral experience of a single incident of violence. This is drawn from official discourses on Self/Other. These discourses essentialise difference and unequal subject positions within China through the minzu category. However, Mr Yan was (re)performing the meanings of these essentialisations in ways which involved articulating all Uyghurs as a violent and inconvertible security threat. The violent nature attributed to Uyghurs justified the violent security threat Mr Yan chose to exert against them. In much the same way as party-state’s attempts to increase security, these popular ethno-nationalist security practices can only decrease security in the region as they make Uyghurs feel increasingly insecure.

These violent security practices perform ethnic boundaries in ways which contest the meaning of Chinese-ness. Mrs Wu, in her early 30s had spent two years away from as a getihu in Ürümchi, exemplified this ethno-nationalism where the very existence of Uyghurs was framed as a threat. On our first meeting, she asked me “do you like Hanguo?”148 This deliberately conflated the Han ethnic group with the Chinese nation (Zhongguo) by merging the ethnonym Han and the character for country guo. Mrs Wu dismissed my suggestion that this could be offensive to other ethnic groups. She said, as a foreign outsider, “You can’t understand”. She became angry that I would not agree with her that “all Uyghurs are bad”. She said “they are all bad. You can’t understand. They all killed us”. I asked about “7-7”, she responded:

Wu: “They killed us first. We must kill them. If they wanted to kill you, you would kill them?”

148 Assuming that I had heard incorrectly, I enquired if she meant South (hánguo). However, the interviewee scorned my Mandarin tones and assured me that she meant “country of the Han” (hànguo).

221

DT: “I would defend myself but I wouldn’t kill innocent people”,

Wu: “There are no innocents, we have to kill them”,

DT: “Who is ‘they’?”

Wu: “Uyghurs”

DT: “But some are innocent. Look at these people going to work, they are just people”,

Wu: “No! They all kill. You don’t understand”.

This conversation continued as Mrs Wu confirmed that she believed all Uyghurs kill people including elderly Uyghurs, policemen, and even the Regional Government Chairman, Nur Bekri. Due to the party-controlled media coverage, Mrs Wu had not heard of the Shaoguan incident or the subsequent protest at People’s Square. The official framings of violence had severely shaped her knowledge and perspective on how identity and security were being practiced in the region. Mrs Wu became exasperated when I suggested that this could be construed as racism saying this was no such thing because that was an “American problem”. She then pointed to a fruit-seller setting up his stall, saying “look at them! They are same as the turban-heads. They are all just turban-heads”. Mrs Wu’s racism involved a personification of the nation and a nationalisation of the person. “You” and “they” were mutually substitutable with entire ethnic groups. She felt physically insecure because of the threat she perceived from the presence of Uyghurs. After the violence of July 2009, feelings of insecurity were easy to understand. However, the racist framings of security produced and were produced by official discourses of identity politics. This racism reflected the securitised ethnic boundaries of nation-building discourses and minzu tuanjie but was used by individuals as obstacles to its fulfilment. Mrs Wu’s dismissal of Uyghurs as “rag-heads” reflected the exclusion of Islam as un-Chinese. This is drawn from pre-CCP discourses on the centrality of “Han” culture (hua) through semantic hybridisation via the party-state’s discourse on the “inside/outside Three Evils”. It is then popularly (re)performed as Han ethno-nationalism where hua, Han, and China are synonymous.

Those in positions of public employment stated the need for tuanjie. However, they repeated the party-state’s socio-spatialisations discussed in chapter 3 where Uyghur-populated areas are associated with terrorism and danger. For example, Miss Lan who explained religion would disappear with scientific development, repeated the backward/modern binarism but in terms of threat and security: “In Ürümchi, it’s not so severe but in southern Xinjiang because there are so many minorities living together it is more so”. Miss Lan was framing Uyghurs who live in areas with less Han and practice Islam as “severe” thus linked to the “extremism” of “The Three Evils”. Han-populated areas are thus relationally understood as more secure and more Chinese than Uyghur regions. Getihu were less tolerant and more vocal about threats from ‘outside’. Like many getihu, Mr Qiang from chapter 6 had explicitly referred to Uyghurs as “foreign” and

222

“not real Chinese”. Mr Yi, the 40 year old male getihu who said Islam was “all about killing people” told me “it’s just like Afghanistan and Iraq here – crazy. Uyghurs are exactly the same as Iraqis and Afghans”. Mr Hou, a late 40s male getihu who was born and bred in Ürümchi told me “Uyghurs are just like Turks, like – same languages, same religion” before immediately turning to the topic of “7-5” and violence. This (re)performed Han-ness as an ethnic nation against Uyghur-ness, understood as what Winichakul (1994) called the “constitutive outside”.

The party-state’s education programmes and television broadcasts, which laid blame on Rebiya and “the inside/outside Three Evils”, were shaping popular nationalism and providing it with vocabulary to articulate alternative visions of nationhood. We cannot know if people would make such links between dangers outside China’s borders and enemies within without the party-state’s explanation of the event as a threat to Chinese identity-security. However, the direct organisational link between the international “anti-Chinese chorus” and the nominally domestic “East Turkestan” problem was one the party-state had immediately announced on the 5th July and heavily propagated since. Mr Chen, who had blamed the violence on Uyghur separatists, told me “It’s a political problem. Political. They all want to separate from China. Life is better and better in Xinjiang but they still want to separate. The big problem is international. Rebiya and “East Turkestan” are all abroad. They stir things up and organise trouble. They aren’t even here yet they make people kill children and ordinary people (laobaixing)”. Like other interviewees, Mr Chen’s perspective was being mediated through official discourse in an attempt to make sense of seemingly senseless violence.

Popular Han nationalism in the city identified Uyghurs as a “constitutive outside” or significant Other. Uyghurs as a group were being framed as an external threat linked to the “terrorist” Islamic world. These Self/Other-identifications were understood through but also challenged the discourse of the “inside/outside Three Evils”. One 55 year old local-born, getihu woman said “7- 5 was exactly like Iraq”. When news of suicide bombs in Pakistan reached China, Mr Qi, the local-born early 20s getihu who told me Uyghurs were known as “turban-heads”, said “they have so many terrorists in Pakistan. It’s exactly what 7-5 was like. Their violence is terrible. It’s better now but I worry it will be like Pakistan again”. Mr Hu, who had been involved in the violence on the 7th, said that “the 5th was just like Iraq or Afghanistan, all the Muslims out of control with big sticks, causing chaos. The Uyghurs are such a disgusting minzu but then we fought back and it was ok”. He was turning the discourse of “The Three Evils” against Uyghurs as a group, which justified violence against them as nominal members of the same multi-ethnic nation. Mr Yu, a quietly spoken 40 year old taxi driver from had escaped from rioters but saw a young child killed in his car said “it was just like Iraq and Afghanistan. They wanted to turn our country into Iraq and Afghanistan but we are not willing”. Mr Yan, who had to rescue

223 his daughter from Uyghur rioters, explained: “it was a nightmare; it was like Afghanistan and then us, the Americans, fighting the Taliban (laughs)”. Visions of international relations were being projected onto political conflicts ordinarily thought of as ‘domestic’. The party-state was attempting to include Uyghurs as Chinese by delinking their self-identification from Turkic-ness and Islam. However, Han were (re)performing this discourse by excluding Uyghurs altogether as foreign and linked to the international and the foreign outside of community. The party’s discourses of danger which aimed to transform Uyghur-ness were being turned against themselves by Han. Han (re)performed Uyghur-ness as a cultural category, which could not be converted. Uyghurs were thus being performed an inside/outside security threat to the Han nation. In Xinjiang, discourses of Chinese nationalism from above and below are being shaped by one another in both antagonistic and mutually reinforcing ways. Power and resistance are mutually reinforcing practices which cannot ‘control’ one another.

7.1.3: Saving the Nation: Ethnic Nationalism in Ürümchi

There is a tendency amongst both Uyghurs and scholars researching Uyghur identity to view Han and the Chinese state as synonymous. It is not uncommon to read references to “Han rule” in literature on Xinjiang (Bovingdon, 2004, p.5). It is not inaccurate to say that “Xinjiang’s government institutions have been heavily colonised by Hans, and have been subordinated at all levels to the heavily Han party structure” (Bovingdon, 2004, p.4). Party leaders in Xinjiang have all been Han since the establishment of the Xinjiang Uyghur Autonomous Region and the regional government is subordinate to the party in Beijing. Official nationalism mobilises symbols associated with Han and objectivises Han-ness through Fei Xiaotong’s (1988) idea of the “Han nucleus” of China. It is fair to call the party-state’s nationalism “Han favouritism” because it is centres a vision of Han-ness through which other groups can be converted. The same can be said of media displays of bodies of dead Han and not Uyghurs during July 2009 (Smith Finley, 2011, p.75-77). However, if we wish to frame ethnicity and identity through contingent processes rather than timeless essences, then celebrating the essentialisms insisted on by the people we are studying is not much of an advance (Amit, 2002, p.46). Han are not to be found at the centre of China, it is a discourse of what it means to be Han.

Saving the nation was indeed a theme which emerged in official discourse and popular Han nationalism at this time. However, whose nation and whose interests were being served by the nation are questions which as we have seen are heavily contested. These Han (re)performances of “discourses of danger” show how even majority groups, with whom the state appears to be in alliance, turn the discourse of the state against itself in alternative articulations of self- identification. These (re)performances were securitising ethnic boundaries so that securitisation discourses were being turned against tuanjie and for a Han nation. When I asked Mrs Wu if

224 there was a contradiction between saying Xinjiang has always been part of the Chinese nation, a self-defined group, and saying Uyghurs are not part of your nation, she laughed and said “no, there is no contradiction! This place is ours and these are bad people”. Mrs Wu was articulating boundaries of Han-ness and its position as the superior, dominant minzu through which China should be understood. Han-ness was being framed as security by contrast to the boundary- transgressing threat of minzu tuanjie, which sought to include Uyghurs in a new multi-ethnic Chinese community.

When Mr Hu said “we fought back”, “we” did not mean the government. “We” referred to Han nationalists who viewed violence by Uyghurs as a physical security threat and an identity- security threat to the superiority of the Han, which the government would not secure. These unofficial identity-security framings informed the protests of September 3rd discussed in chapter 5. Mrs Wen grinned and raised her fist when she told me there was a “big protest” against the Uyghurs and the government on the 3rd September. Mr Hou had also been involved in the protests. He elaborated further: “it’s a political problem. Nothing to do with the ordinary people…Wang Lequan kill kill kill 149 (raises fist)! There was no response. They did nothing. Wang Lequan is useless. He only looks out for himself, so Xinjiang is still really poor, compared to the rest of China. He does nothing, now we got rid of him…but things are not finished”. This articulated internal boundaries within China between the interests of the people and the government but also between prospering neidi and underdeveloped Xinjiang. Mr Hou was implicitly challenging that he as a working class Han was the centre of China. He was articulating his exclusion from politics.

Middle class interviewees talked around the protests but other getihu highlighted how Han-ness could only be secured by the Han and not by the party-state. Mr Yi, the 40 year old getihu, said “the government failed because they didn’t order the police to kill them so the police were afraid, didn’t know what to do”. Mr Xin, the taxi driver, told me “I was here during 7-5 but I dare not speak the truth. The party dares not and the party won’t let us. Shaoshu Minzu were killing everyone. In every other country they would respond immediately and eliminate them but not here. So we had to defend ourselves and on 7-7 we responded. We gave them a beating (laughs proudly)”. This was a (re)performance of Han-ness through insecurity in Xinjiang and the threat of minzu tuanjie. Mr Xin articulated security through ethno-centrism. His preferred solution to security problems was that of the Republican-era revolutionaries; to “eliminate” Uyghurs, a distinct ethno-national group, rather than assimilate them. Mr Yu similarly framed Han as insecure and the party-state incapable of providing security:

The government did nothing, so 7-7 came and we showed them. We got together and we beat them (points south and grins)…every minzu wants their own independent

149 This was a reference to the chant at the September 2009 protest discussed in chapter 5.

225

country - I understand that. They can protest against the government or a policy. I don’t care. It’s no problem but when they kill ordinary people, we are not willing, we will oppose them.

Mr Yu admitted he had turned to violence to protect “his family and the Han”. “We” was being relationally produced against the “they” that is the ethnic Uyghur Other. This we was not Zhonghua Minzu but Hanzu. He claimed he had “no problem” with Uyghurs protesting for an independent nation, which every minzu wants. Mr Yu performed minzu and nation in abstract as coterminous. This implicitly framed China as an independent nation of the Han. The nation was in a state of insecurity not because Uyghurs wanted their own nation. It was insecure because Uyghurs challenged Han superiority within China through violence, which made Han feel insecure on a daily basis. Mr Yan was traumatised by the injuries to his daughter. He then used ethnicised framings of security to explain how it was “human nature” that if “they” attack your family, “you have to fight back and kill them”:

So on 7-7 we Han got together and attacked them. We had to but we were stopped by the special police. It’s a disgrace we were stopped. They killed our families. My daughter’s wounds are healed but our hearts will never heal, they will always hurt. Everything seems peaceful now as you can see, but in our minds the hurt and the hate will always be there. There is nothing we can do. The young people just need to continue working. But we will organise resistance. We will always be ready the next time, we will organise.

Organising meant relying on existing social networks and telecommunications much as Uyghurs did to organise the July 5th protest. Mr Yan’s response to organise Han residents (“we”) to kill Uyghurs (“them”) (re)produced ethnic boundaries. Security of the body was being framed through identity politics, which understood minzu as the boundary of belonging. The party-state was seen as an obstacle to the security of Han-ness, which needed to be resisted as much as the threat of Uyghurs. Han-ness could only be secured by ordinary Han through the elimination of the purported threat presented by the existence of Uyghurs. Minzu and security were as coterminous as minzu and nation so that identity-security could only be achieved through the minzu-isation of the nation.

Interview materials presented here and the protests analysed in chapter 5 suggest Han-ness cannot be understood solely through the essentialisms in the identity politics of the party-state. There were no calls to overthrow the CCP. However, their framings of security and identity politics suggest no straight-forward alignment with the discourses of the party-state. Hanzu should thus be viewed as a contextual and relational category through which individuals contest both their self-identifications and their social positions. For example, discourses of minzu are less explicit in most other parts of China. For example, Beijing Han would ask me “what are Xinjiang Han like?” This section has shown how the party-state’s model of multi-ethnic nation- building is failing to convert Han in Xinjiang to this multicultural form of belonging.

226

Ethnocentrism speaks to Han nationalists in Xinjiang through the idea of the Han as the “nucleus”. However, it fails to appeal to them because they understand their nation through lineage and language. Han-ness is thus (re)performed not as the nucleus but as the very meaning of their nation. There are “multiple, relational degrees of subalternity” where some groups “distinguish themselves from and speak for those ‘below’, while allying themselves with and speaking to those above” (Hershatter, 1993, p.111). Han individuals invert hegemonic texts against themselves and maintain positions as speakers of Han-ness whilst repeating and mediating the identity politics and social practices of the party-state. The party-state now faces the problem in Xinjiang that by promoting nationalism and its power to identify, it has mobilised Han nationalists. However, it also excludes them because they wish to maintain ethnic boundaries rather than fuse them into a multi-ethnic China. The party-state has given them a vocabulary with which to exclude Uyghurs, who as we shall now see respond with securitisations of their own.

7.2: Counter-Securitising Uyghur-ness, (Re)Performing China

This section will challenge Dru Gladney’s “ethnogenesis” argument that Uyghur self- identifications are produced from above. It continues the approach to ethnicity as just one option within an array of identity politics. Political institutions above and members of nominal groups below shape and set limits to each other’s capacity to articulate these discourses. Uyghurs in China and across Central Asia “feel like minorities” because they understand themselves as a stateless group (Roberts, 2007, p.204). Uyghurs are not as susceptible to the hegemonic ideals of any particular state-promoted identity. However, they have been influenced by both Soviet and PRC civic ideals (Roberts, 2007, p.204; Brophy, 2011). Uyghurs are particularly dismissive of the party-state’s interpellations. Uyghurs often appeal to difference by any means to show they don’t want the model of inclusion on offer and thus to resist hegemonic discourses of nation-building. However, both Han and Uyghurs in Xinjiang profoundly influence one another through the ongoing (re)production of ethnic boundaries in Xinjiang. This section explores how Uyghurs in Ürümchi respond to and (re)perform official party-state discourses and unofficial Han framings of what it means to be Uyghur through securitisations from below. The analysis is divided into four sub-sections, which explore the masculinisation of Uyghur ethnicity, the securitisation of space, minzu tuanjie, and language use as a boundary between Uyghurs.

7.2.1: The Uyghur Masculinisation of Ethnicity

227

Uyghur men in Ürümchi tended to be less concerned with physical security than Han. Mukhtar, the “young urban male intellectual” from chapter 6, played down his fears of violence when I asked him about July 7th. He recounted a popular story that many young men told, in which hundreds of Han had marched to Erdaoqiao/Döngköwrük but had been fought off by only a few dozen Uyghur men. This was in Mukhtar’s words because Han “can’t fight, they are all afraid of blood”. Mukhtar was now a relatively studious and responsible citizen but he laughed telling me how, as a youngster, he used to bully and beat Han Chinese teenagers. Mukhtar, like many young Uyghur men, frequently described young Han Chinese men as feminine because they would “carry their girlfriend’s handbags” and “can’t take being teased”. Fear was asserted as characteristic of the Han community by Mukhtar. Young Uyghur men object to the essentialisations they endure. They also find ways to enjoy the stereotype that they are dangerous because it highlights both their exclusion and their burgeoning masculinity. Young Uyghur men perform masculine bravado as a way to (re)perform Uyghur-ness. This inverts the official text where Han are the active, masculine “frontier-builders” and shaoshu minzu are its passive and feminine recipients.

Jay Dautcher’s (2009) Down a Narrow Road showed how performances of masculinity take centre-stage in everyday Uyghur social practices, particularly in jokes and proverbs used to explain social life. However, these masculinities take on new meanings when they interact with the identity politics of the party-state. Processes of nation-building which place Uyghurs in a “feminine” position of receiving “masculine” Han migrants are framed as threats to the virility of the Uyghur nation. Aynür explained that Uyghurs were from a “glorious Turkic civilisation”, then emphasised how this continuity has been broken by Chinese migration and the idea of Zhonghua Minzu. This reflects concerns amongst broader Uyghur social circles discussed in chapter 6 that China is a threat to Turkic-ness. For example, Uyghur “twilight” (Gunnga) poetry, tends to represent a glorious Turkic past, now lying in ruins and giving way to a gloomy present (Friederich, 2007, p.95). However, Aynür exemplified how this transformation is gendered by explaining it thus: “Uyghur men no longer work in fields and no longer have big muscles. Now they have to work in Chinese offices and their muscles are very small”. China is thus framed as a threat to the sovereignty and the masculinity of the Uyghur nation. The dominance of China as a feminine outsider represents humiliation and emasculation. This foments resistance through re-assertions of masculinity and the ability to inflict and endure violence.

Mahigül, a very slight student in early 20s said that she always felt safe in the city “I am not afraid of them, they are afraid of us”. She said though that “for girls” it was not safe at night because they could be raped. Mahigül rarely left the Uyghur district. She did not fear Han. She was highlighting the danger of sexualised violence by Uyghur men. Mahigül, like Aynür, framed boundaries through gender. She said “they don’t know about Uyghurs because they are afraid,

228 it’s stupid, but we are not afraid of them. They are afraid of our men carrying knives but this does not exist at all”. The bravado shown by Han getihu (“we showed them”) after the violence of July 7th was in many ways a response to the stereotype that Uyghurs are a threat to Han masculinity. Han getihu thus expressed pride at their performances of violent dominance over Uyghurs as an ethnic group. Not all young men were as confident as Mukhtar and some young male interviewees expressed that “for the first time, we were afraid of them”. During this period most people of all ethnicities interviewed expressed that they were afraid to leave their homes. After the event, however, it was hard to gauge if much fear remained as the Uyghur district again became bustling at night at least with Uyghurs.

7.2.2: Space and Security Amongst Uyghurs in Ürümchi

Riots and violence broke out in Bishkek during March 2010, dubbed by international media as “ethnic violence”. Aynür, having seemingly forgotten the violence of July, told me “Kyrgyzstan is such a dangerous place”. I laughed and reminded her that people might say that of Xinjiang after July 2009. Her response was “that’s different I am from here, I know it”. This indicated how discourses of danger are intimately related to identity politics because the inside and outside of community are often framed through securitisation. For Uyghurs, Xinjiang is a safe place because “they know it”. It is outside Xinjiang which offers unfamiliarity and danger as discussed in the previous chapter. However, for the party-state and many Han Chinese migrants the unfamiliarity presented by Xinjiang enables it to be demarcated as a site of danger to be dominated in the relational production of a safe neidi.

Chapter 4 discussed how Uyghur rights groups abroad frame security and danger in ways which articulate the inside/outside in alternative ways to the party-state. Uyghurs are seeking political asylum abroad because they frame outside China through security. Nevertheless, the marginalised and quarantined position of Uyghurs in global politics is exemplified through the party-state’s international relations which enable deportations of Uyghur asylum seekers from other parts of Asia. Han residents saw the party-state as not doing enough to provide security during July 2009. However, Uyghurs frame it as a direct threat to their security as it restricts their international mobility. Several businessmen in Ürümchi, Kashgar, and Korla all complained that their passports were kept by the Public Security Bureau150. This they said was to “stop us leaving China” and to “stop us doing business”, which threatened their livelihood. Many informants explained that as a Uyghur the additional cost of applying for a passport was a bribe in the region of 30,000 RMB as well as taking many influential people for many dinners.

150 Since conducting this fieldwork, ethnic discrimination in the passport application process has become a more visible problem in Xinjiang through the case of Atika, a Uyghur student whose application was rejected with no explanation given by authorities (For example, see: Radio Free Asia (2012e) and Tohti (2012b)).

229

Abdullah wished to go to Turkey but after discovering the additional costs and that he wouldn’t be allowed to keep the passport on his person, he said:

It is so hard for us Uyghurs to go abroad because it so hard to get a passport. It is nearly impossible for us to leave. We are fine in Xinjiang but we can’t go anywhere else. For Han, it’s no problem. Our identity cards say what minzu we are and as soon as they see Uyghur, you can’t do anything. You have to go home.

Enclosure into China and into the category of Uyghur as a Chinese minzu was a threat to someone who wished to do business abroad. After July 2009, Abdullah, Aynür, Mukhtar, and Mahigül all said they wanted to go abroad but they all said it would be impossible because they were Uyghur. They all felt insecure because of the state and its identity-security practices. They felt the party-state was restricting the mobility of Uyghurs to cross external boundaries because of internal minzu boundaries.

7.2.3: Minzu Tuanjie and Security Amongst Uyghurs in Ürümchi

Aynür was called into work on the 7th July and she said she shook with fear for the whole journey. This was because the passengers on the bus were “all Han” and were “all armed with sticks and clubs”. She explained that this was particularly frightening by ethnicising security and because the state would not provide security for Uyghurs: “no one would help me. The police, the government, they don’t care if Uyghurs are killed”. Alongside performances of masculinity, many Uyghurs were frightened by the violence of July. The state was, at best, an observer and at worst, the cause of insecurity. Like Han, Uyghurs imagined security as something self- contained within the boundaries of their ethno-national community.

Uyghurs also felt insecure because of the ways violence had been officially imbued with exclusivist Chinese nationalism. After the publication of casualty figures by ethnicity following July 2009 Aynür said she was “very angry” and asked “are we not all human? What difference does it make what ethnic group we are? This does not help minzu tuanjie”. In this and many other conversations Aynür suggested that common humanity should supersede ethnicity. However, she would then draw essentialised and ethnicised boundaries by claiming the Han “are just like this” and Uyghurs “all face this” when referring to individual instances of discrimination faced by individual Uyghurs. Uyghurs are (re)performing the security discourses of the party-state to articulate a position of exclusion from security within China and from mobility across the world. China thus becomes a danger from the outside and the obstacle to minzu tuanjie. The discourses which stereotype Uyghurs as dangerous are (re)performed in masculine, nationalist bravado. However, these discourses are also turned against themselves. Uyghurs lament the denial of their “human”-ness and their articulation through danger. They

230 use these to challenge the idea of China as a multi-ethnic community and suggest Uyghurs would be more secure ‘outside’ China.

The perceived contradictions between the imagined community of Zhonghua Minzu and the visceral daily practices of an ethnically organised social order were highlighted frequently by Uyghur interviewees. After July, bag searches on buses and all public buildings were implemented seamlessly across the city without notice and without visible challenge. Aynür was searched by Han security guards while bringing a box of fruit onto a public bus. Han with similar luggage were not being questioned. This ethnicised pattern was observed several times on buses. It should be noted though that most of those conducting searches were not security officials but public volunteers, almost all of whom were Han151. These visceral interactions and the impact on Aynür’s own life led her to question whether the party’s conceptual community had any substance. She then foregrounded ethnicity in how she understood herself:

They think I have a bomb! It made me so angry, as if I am a terrorist, as if we are all terrorists. They all just think we are all terrorists. What are we to do? They all talk about terrorism but what can we do? It is impossible for us Uyghurs.

When Aynür felt she was being treated as a “terrorist” rather than a fellow human because of her ethnicity, she questioned her membership of community and the validity of minzu tuanjie. This was not simply a conflict between Uyghurs and the state. Ordinary Han volunteers were doing the security of the party-state. “They”, the Other, came to be understood as the party- state and ordinary Han residents in alliance against Uyghurs. Aynür’s personal experiences of ethnicisation became nationalised as those of the Uyghur nation and the Uyghur nation became personalised through her experiences: “so my situation and you know this is now the situation for the Uyghur people, it is all bad. I just want to leave this place, go as far away as possible. It is terrible here. I want to go far away to see the world, the outside world”. Ethnic boundaries were superseding Zhonghua Minzu and eclipsing Aynür’s hope of being seen as human. Aynür wanted to leave where she felt she belonged not because of difference but because of how difference (i.e. ethnicity) was being socially organised. One of the first times I met Aynür, she was watching national military parades from Beijing on CCTV and she did not understand why I would object to the militaristic displays of nationalism. One of the last times we met, she told me “now, I hate the country”. Aynür had always been proud of being Uyghur and her identification with ethnicity had not been produced by politics. However, what it meant to be Uyghur was being shaped through the party-state’s identity politics which placed the Han at the centre and violence against Han and Uyghurs at different levels of securitisation. The multi-

151 As a foreigner outside these ethnicised conflicts, one was often afforded the luxury of not being searched. Joking with security staff “that I’m not a terrorist, don’t worry” even brought some laughs from the normally stern security guards.

231 culturalism espoused by the party-state was being resisted because it was seen as an ethnocentric gloss to the visceral experiences of discrimination.

All Uyghurs interviewed like Han getihu expressed little hope for minzu tuanjie with several repeating the expression heard amongst Han that “it’s just a stupid political slogan”. One male mid-20s graduated office worker said “the biggest obstacle is that people, different minzu, are not equal”. This inequality is conceived both in economic terms but also in terms of culture or status as Aynür’s demand to be treated as a human attests. When discussing tuanjie, Ahmetjan, a 45 year old Ürümchi-lik illegal taxi driver, said “very few learn any Uyghur language. They are like this. They are the majority, there are so many of them, so they think they don’t have to learn our language and our customs but we just have to learn theirs. They propagate their own thought”. Ahmetjan was resisting the objectivisation of Han-ness in the minzu tuanjie concept by suggesting Han unconsciously “propagate” their identity through social practices. He was also resisting nation-building by framing the practice of Han-ness as inherently assimilations. Aynür similarly objected to this objectivisation. When discussing minzu tuanjie she brought up the concept minzu spirit (minzu jingshen), saying “it means just the Hanzu spirit for every minzu”. Mukhtar told me “I have heard this stupid “ittipaq yakhshi” (“unity is good”) shit all my life, I think every single day but it’s stupid”. For Mukhtar, minzu tuanjie was “temporary” because “in 50 years or so there will be only one culture, Chinese culture”. When asked if minzu tuanjie was possible, Mukhtar laughed:

With Han Chinese people? With the Chinese Communist Party? I don’t think so. After what they have done, it’s hard. They really should make up for it. We sacrificed the way we lived for centuries. We are sacrificing our culture to live here. They should at least do something but of course they don’t realise that. We are not losing our ignorance, we are losing our culture. A lot of Chinese people think that we are not well educated, let them be more modern. But they don’t know what they are doing. They are assimilating us.

For Mukhtar, minzu tuanjie was a temporary measure to maintain stability before the “assimilation” 152 of Uyghurs could take place. Mukhtar was performing the boundaries of community through ethnicity. He was (re)performing identity by inverting the text of minzu tuanjie against itself. Tuanjie became understood as assimilation into an unequal China. Minzu tuanjie as offered by the party-state was impossible in Mukhtar’s eyes because it presented a threat to identity-security. Uyghur-ness as an ethno-linguistic community was the referent. Mukhtar’s security concerns had nothing to do with difference itself. He was concerned with the way difference was being organised to facilitate the “assimilation” of his community by the Han.

152 In English Mukhtar used “assimilation”, in Chinese tonghua (to assimilate), and in Uyghur khänzulishish (“to be made Han”).

232

In chapter 6, Mukhtar, like Mahigül, Aynür, and Abdullah had all defined community as ethno- linguistic. Being Uyghur was about being able to use the Uyghur language. For Mukhtar, the only way to reverse what he saw as the threat of Chinese-ness was through education using Uyghur as the medium of instruction: “That would be a way to slow it down but no, we can’t totally preserve it. I can’t think of how to preserve it unless we get these people (Han Chinese) out of here or let these people start to learn Uyghur”. Mukhtar reversed the hierarchy of the party’s state’s vision of Xinjiang as a frontier of China, with the Han at the centre. He positioned Uyghur not simply as a language for use in the home or even confined to his ethnic group. He insisted that Han Chinese ought to study Uyghur in the same way he had had to study Mandarin, as a national language for public use.

Uyghurs are framing social problems in Xinjiang through the relational securitisation of Han and the party-state as a singular threatening Other. Mukhtar was interested in making friends from all over the world but “just not them. I don’t hate them. I only hate the Chinese government. But since most of them are so brainwashed I can’t imagine being with one. As humans we should have doubts about all things. I just don’t understand this blind faith they have to the communist party. That’s brainwashed”. Mukhtar like many Uyghurs recounted that the CCP were men who thought they were God and thus could command loyalty which no “imperfect beings” should command. Prioritising Chinese language in education contradicted notions of equality. Mukhtar understood himself as part of a “dying nation” surrounded by “brainwashed” Han in alliance with the party-state dominating the region’s resources and jobs. To Mukhtar, language policies were designed to assimilate Uyghurs into Zhonghua Minzu. He then said ordinary Han people promoted this transformative process by migrating to Xinjiang and speaking Mandarin. Aynür and several other interviewees would whisper these narratives, sometimes stopping to check the rooms of the house and the doorway to their homes, saying “even the walls have ears”. Fear of being overheard by one’s Han neighbours or criticising them for being “brainwashed” frames Han and the party-state as one fused imagined community of power constantly guarding a “dying nation”. Joanne Smith Finley (2002) argued that the self- selected vessels through which Uyghurs demarcate ethnic boundaries are not directed against the government but towards Han migrants. However, here we see through the narratives of Aynür and Mukhtar how Han and China, people and government are being fused as part of a singular threatening Other. “People-centred” power produces “people-centred” resistance. This Othering does not reflect the contestation which takes place within the category of Han discussed in the previous section. It stems from the experiences of how difference is organised in Xinjiang in the politics of the everyday.

Minzu tuanjie is framed by Uyghurs as a political project with little relevance to daily practices. When I expressed surprise to Aynür that minzu tuanjie textbooks had no information on

233 different cultural practices, she said “of course, children here don’t study that, just politics. They don’t care about people”. Aynür was challenging the “people-centred” approach by articulating herself as a person and the party-state as removed from the people. She had to teach minzu tuanjie but privately expressed her frustration at having to endorse its views. She said she would formalistically “regurgitate the book” without believing any of it:

You have to write self-evaluation essays- I love the communist party, the future will be bright under the communist party, blah blah blah. It’s stupid, they can get in your brain but they cannot get into your heart, you know it’s all wrong.

She explained that life in Xinjiang was “just like this” and that keeping your job depended on negotiating these discourses, which she described as “fascist” and “extremist”. She discussed “The Three Evils”, complaining how the idea did not allow for any self-identification between two extremes of West and East or Chinese and terrorist. She drew the symbol for yin-yang and said “you see, this is because Han culture understands the world and all things in terms of opposites- hard/soft, high/low, yin/yang’, there is no in-between and no neutral”. As someone who was self-consciously “between cultures”, the “life or death struggle” for minzu tuanjie presented a threat to her own identity and to Uyghurs who understood themselves as Turkic between China and the “West”. She was (re)performing Turkic-ness and Uyghur-ness through neutrality and accommodation through the inversion of the discourse of “The Three Evils”. “Han culture” was thus framed as a threat because its impulse to dichotomise did not permit “between-ness” or multi-culturalism. Like most Uyghurs, Aynür doubted the existence of “The Three Evils”. She saw it as a discourse which unfairly represented Uyghurs as dangerous and a threat to national unity:

How can we have tuanjie when they think we are all terrorists? It’s obvious there is no tuanjie; otherwise you wouldn’t see slogans everywhere. When we see these we know there is no tuanjie. We have to study it night and day, tuanjie this and tuanjie that. One day I was so depressed, studying this all day in classes. It was getting dark and we were all so tired. Then I left the class and my mood was lifted because I finally found an advert for a new house. So I phoned to ask if they still had the house. He said yes but then asked me what minzu I am. I told him I am Uyghur. He said he was not willing to rent me the house. I asked him why and he said ‘you are a Uyghur, there is no other reason’. So I feel minzu tuanjie is like pressing my warm cheek against their cold ass.

Aynür was expressing her resentment that she felt she had attempted to participate in minzu tuanjie by interacting with Han Chinese residents but was continually rejected as unwelcome in her homeland. She used a Chinese idiom which refers to the lack of appreciation for a host’s hospitality shown by guests, reminiscent of Abdurehim Heyit’s “stubborn guest” mentioned in chapter 6. Aynür implicitly drew boundaries between herself as indigenous to Xinjiang as a form of chthonic self-identification and the Han as “guests” in her homeland. However, her “warm cheek” represents her own individual willingness to be welcoming and to be part of a community with the Han. This hospitality (mehman dost) was being performed by her

234 willingness to use the Mandarin language and the Chinese cultural idiom itself. “Their cold ass” represents the inhospitability of the Han as a group, despite her attempts to integrate and their claims to seek tuanjie. This was a complaint that the Han as an essentialised group affected her as an individual in leading her daily life; her warm cheek against their cold ass. For Aynür minzu tuanjie was part of a conceptual apparatus that simply did not relate to her daily visceral experiences of ethnic discrimination. A group of Han laughed at Aynür when she interrupted to tell them Erdaoqiao was part of B-qu not D-qu: “it’s always just like this, unity is only superficial”. The more her ethnicity as she understood it was interpellated as a threat, the more she securitised ethnic boundaries.

7.2.4: Language as Boundary Between Uyghurs in Ürümchi

The securitisation of ethnic boundaries was not confined to demarcating the external lines between one discrete, impenetrable community defined against another. The indivisibility of the internal and external boundary was being articulated through discourses of assimilation of the internal into the external. Mukhtar articulated his fear of Uyghurs becoming dominated and female by being converted: “The Chinese try to block our vision, make us satisfied with development. The most important thing is we keep our language alive”. Mukhtar felt Uyghurs were being swayed by the promise of development and were “forgetting their culture” by speaking Chinese. The external boundary of urbanised, feminised Chinese-ness was projected internally into the Uyghur community. The internal, Sinicised Other were young Uyghur boys who spoke Mandarin instead of Uyghur and played “Chinese games” such as cards, rather than playing sports and stealing apples from orchards.

One cultural boundary, which was frequently highlighted in interviews and which cuts across the ethnic boundary between Han and Uyghurs is the minkaohan-minkaomin boundary. Minkaohan refers to ethnic minorities who are educated in schools using Mandarin as the medium of instruction. Minkaomin, conversely, denotes ethnic minorities who are educated with Uyghur or other minority languages as the medium. Minkaohan are often referred to as Xinjiang’s 14th minzu because they are seen as distinct from Xinjiang’s 13 ‘indigenous’ groups (Smith Finley, 2007, p.224). Jennifer Taynen (2006) showed that minkaohan are economically advantaged by enjoying job opportunities not available to those whose Mandarin level is not as high. However, minkaohan are also socially isolated because they are more exposed to racism from Han due to easier communication (Taynen, 2006, p.45-46). However, they are also often isolated from other Uyghurs because they are seen as Sinicised and occupying an “ill-defined and uncomfortable middle ground” (Taynen, 2006, p.45-46). Minkaohan are “betwixt and between” because they are often seen as neither wholly Uyghur by Uyghurs nor wholly Han by Han, who refer to them as “half-castes” (erzhuanzi) (Smith Finley, 2007, p.219-220). Mukhtar

235 described minkaohan as full of fear because “they were afraid to make Uyghur friends” due to Chinese influence: “They are different, miserable to be honest. They are just scared, disorientated”. This boundary-demarcation placed minkaohan outside the Uyghur ethno- linguistic community and inside the unnatural, imagined community of Zhonghua Minzu. For Mukhtar the Uyghur language symbolised visceral community and masculinity whereas China represented assimilation and femininity. However, minkaohan represented the outcome of the security threat of China because they transgressed the boundary between visceralised ethnic communities and imagined conceptual communities.

Mahigül, like other interviewees, (re)performed the boundaries of Uyghur-ness through language use by saying that Uyghurs who could only speak Mandarin would become Chinese. However, she expressed different concerns about the level of threat posed by assimilation under the Chinese-state than those of Mukhtar. She securitised the need for Uyghurs themselves to maintain ethno-linguistic boundaries rather than directly resisting Chinese nation- building. This reflected her understanding that Uyghur culture needed to “create new things” and a rejection of forms of Uyghur heritage deemed traditional. Mukhtar stressed the need to maintain a sense of community with the imagined community of the Uyghur dead. However, Mahigül emphasised language was to be used to “progress” regardless of its “centuries of history”. This reflects popular discourses amongst Uyghurs, including that of Abdurehim Heyit, which emphasise tradition and continuity. However, these discourses simultaneously reject the positioning of Uyghur-ness as a frozen relic to be gazed upon and removed from the dynamic change associated with modernity and inter-cultural interaction. When I asked Mahigül if she worried like many other Uyghurs about the future of her language, she said:

If there are some languages that have disappeared, it’s because there are no more people who speak in that language but we are eight million Uyghurs. It’s not possible. It wouldn’t happen that all of us would speak Chinese and forget Uyghur. So still there would be say a million people who could speak Uyghur, so it wouldn’t disappear.

Mahigül’s upbringing in Ürümchi meant she was accustomed to an urban centred on the nuclear family in a gated apartment block. Her community had always centred on an apartment where family would congregate in private with doors closed and locked. This allowed her symbolic security from the boundary transgressing practices of the Chinese party-state and Mandarin speaking minkaohan. For Mahigül, as long as Uyghur could be spoken in the home, it would retain its symbolic power and “wouldn’t disappear”. Mukhtar on the other hand explicitly longed for a return to a community life where “all families hang out together on the street”. This was not because he was raised in a remote rural region. He was accustomed to a different type of urban environment- a Uyghur Mähällä where doors are often left open and there is

236 frequent interaction between neighbours153. When we discussed the preservation of languages it emerged that Mahigül and Mukhtar shared a sense of community, defined with reference to the common symbol of language. However, the meaning of this symbol was not shared. They both explained that it would be laughable to speak any language other than Uyghur with their family and friends, with the exception of foreigners. However, Mahigül rejected the need for public education and took a more individualistic approach. This was again rooted in her experience growing up in a nuclear family in a private apartment in the big city of Ürümchi:

You can maybe speak Chinese at school or at work but at home you have your choice. I don’t think it’s a problem. It’s our job to keep it. It’s not about education. I think it’s first the parents job, not the school. If my kids went to Chinese class, I would teach them Uyghur at home when they are very small.

Although less traditionally political, Mahigül was a fiercer critic of minkaohan than Mukhtar. This was because Mahigül saw language maintenance and the reproduction of community boundaries as the moral responsibility of the individual and the family within this community, rather than the community per se. She was happy to explain to me the purported differences between minkaomin, who she said were more fun to hang out with because they used Uyghur language, and minkaohan who she did not socialise with because they were like Han Chinese. This cross-cutting boundary of we-ness performed through language was linked to how she Mahigül imbued social practices with meaning. She even went so far as to inverse the positive stereotype of Chinese as hard-working to mean “they are boring”. She contrasted this with her own understanding of essentialised Uyghur-ness to explain “we are a lazy minzu, not like the Chinese, we are good at having fun”. This took Chinese stereotypes and inverted them so that Uyghurs being “lazy” meant they were good fun to spend time with. Mahigül turned an external boundary inwards as minkaohan and minkaomin were being defined through the dichotomies of boring/fun and Chinese/Uyghur. Unlike Mukhtar, Mahigül’s essentialisations were not qualified and despite her claimed between-ness she often referred to minkaomin as “we”. According to Mahigül, the fundamental differences between minkaohan and minkaomin were what language they chose to speak and the ethnicity of their friends. She would lament that minkaohan may choose to date and socialise with Han Chinese. She offered little explanation why this was a problem, other than “they are Chinese”. This performed a securitisation of the minkaohan/minkaomin boundary where the real threat to Uyghur-ness was the transgression of the Chinese/Uyghur boundary by the minkaohan internal Other. This placed minkaohan outside her symbolically constructed visceral community and into the imagined community of Zhonghua Minzu:

153 This could be read as a gendered view of politics but other female interviewees from Korla shared the same concerns as Mukhtar for the public status of Uyghur. On Mähällä, see Dautcher (2008) Down a Narrow Road.

237

They are more open-minded but not in a good way…We speak Uyghur. Some minkaohan speak Chinese. Their Uyghur is not good, I don’t like that. The Uyghur classes (minkaomin), they study in Chinese but they communicate in Uyghur and they hang out with Uyghur friends but they (minkaohan) hang out with Chinese. They will be influenced by the Chinese, you know, from primary school to university.

The boundary of community being drawn here reflects the indivisible internal/external boundaries of self/other, minkaohan/minkaomin, and Uyghur-ness/Chinese-ness. Mahigül and Mukhtar imbued Chinese-ness and language with different meanings. Nevertheless, they self- consciously participated in internal/external boundary demarcation. Language and imagined positions in a spectrum of what they themselves called “Sinicisation” were symbolically constructing community. Mahigül said she thought that ethno-centrism was so fierce in Ürümchi that Han Chinese should not learn Uyghur because then they would “know we are cursing them” just like “we know when they curse us”. She said she had no feelings whatsoever whether non-Uyghurs learned Uyghur, including foreigners because it was “our language”. This was unlike Mukhtar who, concerned with the social status of Uyghur, expressed pride when he met several foreigners who at least tried to use basic Uyghur phrases. They were imbuing Uyghur with different meanings (public/private) but they were equally concerned with its preservation as a practice and as a securitised boundary defining their community. Even though Mahigül framed the boundaries of community in very similar ways to Mukhtar, there was no China threat as such; the threat was Chinese-ness as a mode of behaviour. The threat was not China as a political unit but the cultural influence of Chinese-ness which transgressed the boundaries of Uyghur-ness such that minkaohan were the physical threat. There was no China threat because for Mahigül, the Uyghur language symbolised the home, the domestic, and the small-scale community. She would use Mandarin out of necessity and Uyghur out of the love of her mother-tongue and her community. The responsibility to maintain this linguistic community lay in the hands of the individual and the nuclear family. This was counter-securitisation but it was minkaohan who were being framed as a security threat to her sense of we-ness. Resistance was the responsibility of the individual not the collective.

Mahigül laughed when I asked if the categories of minkaohan and minkaomin referred to the school system one had attended and re-iterated that it was about what language one used not what school one attended. They both suggested these categories will disappear as formal, educational designations under Bilingual Education policies. However, they both said they would continue to use them to refer to the language abilities, or the level of assimilation, of other Uyghurs. As Anthony Cohen argued people “think themselves into difference…the appearance of continuity is so compelling that it obscures people’s recognition that the form itself has changed” (Cohen, 1985, p.86 & 91). The boundary may appear rigid but its meaning is contingent and shifts with social practice and the idiosyncrasy of those who employ them. Social categories persist and ethnonyms can be justified through appeals to historical continuity

238 but their meanings are sites of daily performance, negotiation, contestation, and (re)performance. Mahigül told me she contravened the tenets of Islam by drinking alcohol and by wearing dresses which revealed her forearms. These admissions indicated both the contingency and the persistence of these boundaries. She then claimed minkaohan were less Uyghur and less Islamic than minkaomin. She attributed this dress-sense to “Chinese influence” through the medium of minkaohan schooling and mixing with Han Chinese. By wearing clothes which showed their upper arms they contravened Islam more gravely. The contingent boundary between Uyghur Muslim and Sinicisation shifted from wrist to fore-arm: “We are Muslims. Chinese people dress like that, not us”. Minkaohan were being positioned as a threat outside of Mahigül’s self-defined ethno-linguistic community because she saw them as less Uyghur and more Chinese than her family and her minkaomin friends. Language was being understood as productive of Uyghur-ness whereby speaking Chinese would mean becoming Chinese and behaving in a Chinese way. By linking language to these social practices, being Uyghur was then a performance of internal/external boundaries. These performances took place on a daily basis and in banal, seemingly apolitical ways, such as through dress-sense and having fun.

The party-state’s transgression of ethnic boundaries in Xinjiang is producing resistance rather than conversion. Chinese nation-building in Xinjiang has created constant social anxiety about the preservation of Uyghur language and what it means to be Uyghur through the prism of the internal/external boundary of ethno-linguistic community. The contrasting meanings with which interviewees imbued the Uyghur language show how symbols are ideas through which “individuals can express their attachment to a society without compromising their individuality” (Cohen, 1985, p.18). Uyghur language was being mediated through the idiosyncratic experiences of individuals. It nevertheless retained its social power as a shared symbol, which demarcated the boundaries of their community as they imagined it. For example, Mahigül, like Mukhtar wanted to maintain the Uyghur language and saw it as her responsibility to preserve this defining boundary of her community against the existential threat of “Sinicised” minkaohan. However, for her the meaning which she sought to preserve was a domestic language which afforded her and her community privacy from outsiders. She did not seek to reverse stigmatisation or overtly resist assimilation as such, she largely dismissed it. Mahigül wanted to continue living as she had done all her life, in a family unit within an urban environment where two ethnic groups could not speak each others’ language and did not include one another. This would prevent the transgression of the boundaries of her imagined visceral community. Mahigül and Mukhtar personalised and collectivised Uyghur-ness at the same time. They defined themselves as authentically Uyghur but extrapolated their individual experiences through a performative enactment of collective boundaries. These performative enactments permitted the inclusion of diverse individual experiences. However, they excluded Minkaohan as transgressors of the cross-cutting internal/external boundary of Chinese-ness/Uyghur-ness. Mahigül and

239

Mukhtar were not listening to the voices of nation-builders which defined them as “backward”. These voices could not be re-appropriated in ways which made sense to their individual lives. They were listening to the voices of their family and their minkaomin friends to (re)produce the boundaries of Uyghur-ness to exclude Zhonghua Minzu as a threat to their identity-security.

This section has shown how Uyghurs in Ürümchi negotiate and (re)perform official party-state discourses on Uyghur-ness by engaging in counter-securitisations from below. These securitisations are (re)performances of ethnic identities, which are conceptually similar to those of Han which emphasised lineage and language. These (re)performances resist the party-state’s nation-building as a form of assimilation. Assimilation into Han and China are taken to be synonymous because ethnicity is organised in hierarchical ways which conflate the two. China is thus framed as an outside security threat which threatens the gender, the prosperity, and the cultural symbols of the Uyghur nation. The boundaries which encapsulate the Uyghur community cannot be understood from inside or outside but from how the inside and outside intersect. Ethnicity as a discursive category was mediated by the idiosyncratic experience of the individual interviewees thus allowing them to “personalise” nationalism. The contingency of boundaries allows different individuals to imbue different meanings to symbolic categories. However, these categories retained their power to shape the framing of Self/Other and security/danger in ethno-linguistic terms.

Conclusions

Ethnic identity politics and security are interlinked contested discourses. Examining identity through the dynamics of identity contestation, allowed the analysis of the relations between articulations and securitisations of identity from above and below. The official politics of the party-state and everyday self-identifications from below are best understood as inter-linked discourses which shape each other. This chapter showed how Han and Uyghurs (re)perform alternative configurations of Self and Other through the official discourses of identity-security in Xinjiang.

The first section showed how Han in Ürümchi justified violent ethno-centrism through a conceptual apparatus given to them by the multi-ethnic nation-building of the party-state. Liberation was a concept used to place Han-ness in a position of superiority to Uyghur-ness defined through backwardness. This backwardness was mediated through discourses of danger. Backwardness became securitised as a threat to the nation linked to a terrorist, Islamic outside. Amongst Han, the category of Hanzu was also being contested through how the Other was framed. The getihu appealed to ethnic identity based on lineage and language. This suggested

240 the inconvertibility of Uyghurs and the need to violently dominate them. However, the middle classes retain an interest in the attraction of the “Han nucleus” and its ability to convert backwardness into modernity and Uyghurs into China. These are contestations over how to conceptualise the Other. However, they are also contestations of how to frame the Self through its relationship with Otherness. There was little if any unproblematised alignment with the party-state amongst Han. Hanzu is best understood as a contextual and relational category which conceals contestation over the meaning of its boundaries. Han individuals, like Uyghurs, invert hegemonic texts against themselves for their own ends. The party-state’s model of multi- ethnic nation-building places Han at the centre of the nation. This is drawn from older discourses on Han nationalism but gives contemporary Han nationalists the ethno-centric vocabulary to draw from. However, this creates insecurity amongst Han nationalists. They (re)perform their ethnic identity as under threat from being forced to include Uyghurs in their national community.

The second section showed how exclusion and unequal inclusion reinforces ethnic identification amongst Uyghurs. They draw from visceral experiences of ethnicised exclusion to reject that they are being included in any positive ways at all. Uyghurs are turning minzu tuanjie against itself and cast doubt on its supposedly pluralist intentions. The model of purportedly peaceful conversion promoted by the party-state and the Han middle classes presents an identity- security threat for Uyghurs. Minzu tuanjie and a Han-centred Zhonghua Minzu are framed as an assimilationist transgression of ethnic boundaries for the benefit of the Han. Uyghurs in Ürümchi are (re)performing official party-state discourses on what it means to be Uyghur by engaging in counter-securitisations from below. China is framed as an outside security threat which threatens the masculinity, the prosperity, and the cultural symbols of the Uyghur nation. Han and China are taken to be synonymous because ethnicity is organised in hierarchical ways. Minkaohan and Uyghurs who prefer to speak Mandarin symbolise the transgression of these boundaries and are being represented through these discourses of assimilation by Uyghur- speaking Uyghurs. Minkaohan are thus being excluded from Uyghur-ness by Uyghurs who securitise this internal/external boundary. They are framed as being less Uyghur than minkaomin in ways similar to how Uyghurs are officially articulated as less Chinese than Han.

Han and Uyghurs conceptualise threats to their person and their sense of Self in ways that are shaped by official discourse but which simultaneously resist this model of nation-building. Despite overt dismissals of the boundaries articulated in official nation-building discourses, Han and Uyghurs used these discourses as means to resist them and articulate ethnic identities. Both groups used the inclusive concept of minzu tuanjie to exclude the ethnic Other. Daily practices were mediated by ethnic boundaries and highlighted to indicate how the behaviour of the Other contradicted notions of a shared multi-ethnic community. Han and Uyghurs are both

241 contesting the meaning of ‘China’ through the meanings they give to daily practice. Both groups represent the party-state as preferentially treating the Other in ways which allow assimilation or the devaluation of their ethnic identity. The Other in alliance with the party-state is thus framed as a threat to identity-security. Nation-building attempts in Xinjiang are failing to convince both the Han majority and minority groups that they are part of the same national community. They continue to Other each other and choose to define their communities along ethno-linguistic lines.

Uyghurs and Han in Ürümchi tend to show little commitment to the conceptual imagination of Zhonghua Minzu. Instead, they increasingly understand themselves as discrete and impermeable visceral communities. The party-state is attempting to convert difference and convince all the peoples of Xinjiang that they are part of one community. However, it is doing so in a Han-centric way which hierarchically organises difference. This only reinforces the difference it aims to convert. The party-state now faces the problem in Xinjiang that by promoting nationalism and its power to identify, it has mobilised and excluded Han nationalists at the same time. These nationalists wish to maintain ethnic boundaries rather than fuse them. They see themselves as superior and modern, which draws from ideas about the structure of Zhonghua Minzu. The party-state has given them a vocabulary with which to exclude Uyghurs with, who respond with counter-securitisations of their own. Han see Uyghurs as a backward danger to be overcome by conversion or elimination. Uyghurs see Han/China as an identity- security threat aligned with the party-state’s attempts to assimilate them. These securitisations are mutually constitutive but mutually incompatible. The solutions to social problems posited by each side are generally regarded as the basis of the problems by the other. The ethno-centric approach of “Scientific development” is seen as a threat by Uyghurs while protesting against inequality is framed as “terrorism” by the party-state. An (in)security dilemma has emerged where the imposition of the party-state’s framings of identity-security is making people on all sides feel insecure. The party-state does not have a monopoly on security or securitisation practices. If it does not learn how to co-opt or adapt to these demands from below, it can expect increasing insecurity in the region as ethno-centrism is exacerbated by inequality and violence.

242

Conclusion: (Re)Building the Nation

This thesis has asked how we can theorise the relationship between security and identity in IR but it has done so by using micro-fieldwork to explore the relationship between ethnicity and nationalism in Xinjiang. The analysis asked what type of nation is China becoming by asking how the party-state conceptualises and makes policy to manage difference on its northwest ‘periphery’. Discussions of nation-building ought to incorporate perspectives from below as much as those of elites to explore the limits and influence they place on one another. Weber and Lacy (2011) have noted that the pursuit of security is inherently problematic. The destabilising nature of security has particular resonance in the case in Xinjiang where securitisation of Chinese identity-security has been shown to produce insecurity amongst groups who identify themselves in alternative ways. Nation-building is beset by the same fundamental problems that it cannot include everyone and some groups are included in ways they find run counter to how they define themselves. Just as the power of securitisation produces resistance through (re)securitisation, nation-building is faced the inevitability that the nation(s) it interpellates will respond by (re)building the nation through everyday speech and the attributions of meaning given to daily practices. The finality and singularity offered in nation-building cannot produce the boundaries it performs because its discourses are turned against themselves and the state has to respond the efforts from below to (re)build the nation. In the case of China and Xinjiang, the party-state remains beset by Han chauvinism which is drawn from scholars such as Fei Xiaotong and Ma Rong who take Han-ness as a model of conversion for non-Han ethnic groups within the contemporary territorial jurisdiction of the PRC.

The first three chapters largely explored official discourse to explore what ‘China’ means in the party-state’s nation-building project. The first chapter outlined the theoretical and methodological approach of identity-security and explored its application to the empirical cases of China and Xinjiang. This used literature from studies of nationalism and nation-building in China and Xinjiang to argue that nationalism is best understood as a process of identity-security performance from above and below. Chapter 2 showed how the framing of social categories in the writing of history is central to the Chinese party-state’s nation-building project in Xinjiang. Xinjiang is represented as an indivisible component of the Chinese nation but its inclusion in the political community is coupled with exclusion from the Chinese cultural community. Spatial divisions between regions of China (neidi vs. bianjiang) and between ethnic groups (Han vs. Shaoshu Minzu) are employed as boundaries within the nation. These boundaries are based on oppositional and mutually reinforcing framings of self/other, inside/outside, backwardness/modernity, and security/danger. These dichotomies perform a hierarchy of

243 identities where Han-ness is framed as the nucleus of the nation and non-Han identities are peripheralised and await their own extinction (minzu xiaowang). Chapter 3 adapted the concept of securitisation to show that this is a political act of boundary production rather than a reflection of pre-existing, objectively verifiable identities. It showed how securitisation is central to the CCP’s nation-building project in Xinjiang because it demarcates the boundaries of the nation (Zhonghua Minzu) by articulating who and what is a threat to its existence (Turkic-ness). The party-state’s explanation of all incidents of unrest and dissent in Xinjiang through the discourse of “The Three Evils” showed how all resistance to the identity-security offered through Zhonghua Minzu is securitised as terrorism and criminalised as ethnic separatism.

Chapter 4 showed how different meanings were attributed to violence by different groups in ways which order the nation. Violence by Uyghurs during July 2009 was securitised as a threat to China where violence by Han was either represented as an ordinary issue of “public order”. These discourses attached different ethnicised values to different bodies. The different meanings attributed to violence shaped how different groups were treated by the security apparatus, particularly through the practice of arbitrary detentions and illegal deportations from China’s Asian allies such as Kazakhstan and Cambodia. This chapter also analysed how the dichotomies of security/danger and ethnic/national were performed in official discourse and (re)performed in popular Han responses to the violence. Han protests were against the government but for the Han nation in a move which complicates the ordinarily unproblematised relationship between Han and the state. Chapter 5 showed how the model of inclusion offered to Uyghurs (minzu tuanjie) marginalises Uyghur-ness by placing Han-ness at the centre of the nation. Social practices associated with ethnic Han are objectivised as national and inclusive creating an unconsciously ethnocentric nation. The inclusion of minzu tuanjie and the exclusion of “The Three Evils” are inter-textually related and mutually constitutive aspects of the same nation-building “struggle”. Uyghurs are to be converted into nationals but the symbols used to define the nation are understood by Uyghurs and Han alike as symbols of Han ethnicity (e.g. Mandarin Chinese, the Great Wall, and the Yellow Emperor). These two chapters supplemented the first three by exploring how nation-building takes place in the politics of the everyday through education and on the street as much as it does through the training of cadres and the writing of white papers analysed in the first three chapters.

The final two chapters explored how these nation-building performances are perceived and (re)performed by the social groups they seek to unify in a multi-ethnic, Han-led China. Chapter 6 showed how the party-state has been unable to assert a monopoly over the boundaries and meaning of identity in Xinjiang. Most official and popular discourses are alternative modes of essentialisation. These essentialisations attempt to articulate timeless discrete communities in the face of social contingency. The party-state does define who belongs to what minzu but its

244 attempts to define what these categories mean are taken out of their hands by people who use this discourse for their own ends. Han in Ürümchi contested their own identities by how they framed the position of Xinjiang in China. Intellectuals lean towards adopting the concept of minzu tuanjie and the objectivisation of Han-ness this entails to promote nation-building in the region. However, the working class and getihu Han tended to reject it altogether as contradictory to their nation defined through ethno-national terms. Uyghurs imagined a timeless Turkic civilisation running parallel to Chinese civilisation, rather than above or below it. Uyghurs in Ürümchi, like getihu Han, see the model of multi-ethnic community with the Han at the centre as incompatible with the intra-ethnic social practices of their daily lives. Party-state performances of Chinese-ness are having the opposite effect of that intended as they are (re)performed such that Uyghur-ness and Han-ness are enacted through discourses of timelessness, civilisation, and culture.

The final chapter explored how ethnic identity politics and security are interlinked and contested discourses. This chapter showed how Han and Uyghurs (re)perform alternative configurations of Self and Other through the official discourses of identity-security in Xinjiang. Han in Ürümchi justified violent ethno-centrism through the ethno-centric conceptual apparatus of “liberation” and “backwardness” given to them by the multi-ethnic nation-building of the party-state. Liberation was a concept used to place Han-ness in a position of superiority to Uyghur-ness defined through backwardness. This backwardness was mediated through discourses of danger. Backwardness became securitised as a threat to the nation linked to a terrorist, Islamic outside. The getihu appealed to ethnic identity based on lineage and language. This suggested the inconvertibility of Uyghurs and the need to violently dominate them. However, the middle classes retain an interest in the attraction of the “Han nucleus” and its ability to convert backwardness into modernity and Uyghurs into China. However, exclusion and unequal inclusion reinforces ethnic identification amongst Uyghurs. They draw from visceral experiences of ethnicised exclusion to reject the idea that they are being included at all. Uyghurs are turning minzu tuanjie against itself and cast doubt on its supposedly pluralist intentions. Minzu tuanjie and a Han-centred Zhonghua Minzu are framed as an assimilationist transgression of ethnic boundaries for the benefit of the Han. China is framed as an outside security threat which threatens the masculinity, the prosperity, and the cultural symbols of the Uyghur nation. Han and China are taken to be synonymous because ethnicity is organised in hierarchical ways. Identity and security are intimately related in Xinjiang and the micro-fieldwork approach used here was effective because it is showed how security practices create insecurity on the ground and nation-building can produce alternative nationalisms. If it is possible to assess the success of nation-building the answer would be that it is a resounding failure because people in Xinjiang feel increasingly insecure that their nations are under threat.

245

This thesis has shown how Uyghurs and representations of Xinjiang problematise the relationship between ethnicity and nation in China. The unique contribution of the fieldwork was that this was the first ethnographic study of responses to the violence of July 2009. Furthermore, the incorporation of Han perspectives has been very limited in the literature on Xinjiang. Interviewing Han residents of Ürümchi here has given insight into how Uyghurs and Xinjiang are represented in different ways by different groups. This showed how Han perspectives cannot be assumed to be aligned with or a reflection of party-state discourse. ‘China’ then becomes a site of contestation where Uyghurs are Othered in different ways, which articulate different Chinas. Stressing different racial or ethnic origins means from Uyghurs shows that many Han are (re)building their own nation because the party-state’s model of a multi-ethnic community does not correspond with how people understand their daily lives. , a Uyghur associate Professor of Economics at the Central University of Nationalities who is currently under house-arrest for his publications on the blog Uighurbiz, found that 89.4% of Han in Xinjiang wanted the government to maintain the “superior position” of Han in Xinjiang (Tohti, 2012). All surveys in the region are beset by political difficulties but the interviews conducted for this thesis also found that a Han superiority complex is a discourse which many Han in Xinjiang continue to identify with. For example, Mrs Wu in chapter 7 called all Uyghurs “turban-heads” and used this to justify the murdering of Uyghurs became exasperated with the suggestion that this was racism because that was an “American problem”. Mrs Wu was obviously using racist vocabulary and essentialisations but, like minzu tuanjie, her perspective remained undisturbed by the exclusion felt by Uyghurs in the way they are included as inferior. Xinjiang and Uyghurs are thus articulated as belonging to China not in the sense that they are equally integrated into a modern sovereign state but that they remain an imperial possession of China to be Sinicised. A serious obstacle to addressing discrimination is that the hierarchies of identity in Han chauvinism remain concealed by discourses of minzu tuanjie, which as chapter 5 showed, offer unity on the surface and but a concealed hierarchy when analysed more rigorously.

The discourses of Han chauvinism and minzu tuanjie speak to one another but they do not speak to Uyghurs. For example, Ilham Tohti’s survey found 67% of Uyghurs believe the Chinese government is a “representative of Han interests” (Tohti, 2012). The ‘new’ “second generation” debate of minzu policies analysed in chapter 2 offers no voices from Xinjiang. Failing to consult scholars from Xinjiang excludes their expertise on how policies are actually put into practice in the region. However, it also indicates one place where the ‘red line’ of censorship in contemporary China is stark and distinct. Uyghur scholars such as Ilham Tohti have devoted their careers to addressing these issues. In 2009 Ilham Tohti wrote a detailed essay suggesting that periodic episodes of violence in response to policy changes show that these policies need to be reconsidered and debated (Tohti, 2009). Ilham Tohti has since spent much of his life

246 under house arrest for publishing research on the subject. Conversely, a scholar such as Hu Angang, a Han Chinese political economist with limited expertise on ethnicity, nationalism and Xinjiang, writes essays for this debate and is vaunted as one of China’s leading thinkers.

Ironically, Ilham Tohti’s vocabulary and prognosis for Xinjiang’s problems is much the same as that of the party-state: “without the development of ethnic minorities, there can be no revival of the Chinese people…and without dealing with the poverty and backwardness of minorities, it will be hard to maintain long term stability in Xinjiang” (Tohti, 2006). Where Ilham Tohti diverges from the party-line is in his diagnosis that Uyghurs face “systemic discrimination” in “entering the advanced productive forces”, most notably in the resource extraction industries which are the basis of the CCP’s development plans for Xinjiang (Tohti, 2006). The extraction and export of Xinjiang’s natural resources by an almost exclusively Han workforce has long exacerbated the sense amongst Uyghurs that their resources are being “expropriated” (Bovingdon, 2004, p.47). Mr Wang, the Han oil executive interviewed in chapter 5, openly admitted on our first meeting that his office in the state-owned SINOPEC, the largest oil extraction company in Asia, refused to employ any Uyghurs because “none of them work, they don’t study. We used to have one but now I won’t employ them as they are too much trouble. Too much arguing. So many disagreements and ethnic contradictions”. When pressed, Mr Wang explained that such “ethnic contradictions” were that Uyghurs wanted to pray and wouldn’t eat pork. Continuing this model of development without addressing the discrimination which the likes of Mr Wang maintain will only further marginalise Uyghurs. However, what is most significant here is that debates on minzu policy have been ongoing for a long time in Xinjiang amongst scholars and amongst the broader populace as chapters 6 and 7 have shown. The party-state’s attempts at a more open debate are merely a response to these social dynamics which it cannot control. The party-state wants to regain control of the minzu discourse in a way which permits a greater degree of open-ness but focuses the debate on the banal “salad bowl” versus “melting-pot” alternatives. Narrowing the debate in this way enables the exclusion of serious discussion by experts on the subject. The of scholars who address social relations and discrimination in the political economy of Xinjiang indicates one red line in China which is non-negotiable and should not be openly discussed: ethnic discrimination.

The renowned Confucian scholar, Tu Weiming (2007), when comparing “convergence and divergence of core values between American and Chinese civilizations” urged the world to look beyond the values of the Enlightenment by proclaiming “empathy is at least as important as rationality”. Tu Weiming offers a useful way of looking beyond any “clash of civilisations” by offering a practical means of at least accommodating Otherness through empathy. However, who is deemed worthy of this empathy in international politics is a question of power. Empathy should not be restricted to the geopolitical debates on China and the US, which say little about

247 the rest of the world or competing perspectives within these nations. The category of China demands that empathy be denied to those social groups within the Chinese state who see themselves and have been historically labelled as civilisations outside of Chinese culture (hua). One local born Han woman told me “we can’t be real friends…our hearts will always hurt”. Mr Yan, the taxi driver who rescued his daughter from the violence explained “there is nothing we can do, that’s it for a generation, my generation hated them and my children’s generation will because of what they’ve done”. The fact that violence was immediately framed through the essentialising identity politics of “we” and “they” show that problems of inclusion and exclusion in Xinjiang are neither new nor on the verge of any kind of ‘resolution’.

If the party-state and indeed Han nationalists were able to apply empathy to the politics of regions such as Xinjiang instead of hatred and silencing, security would be considerably improved because Uyghurs would not feel threatened by ‘assimilation’. For example, the young businessman, Mukhtar, explained the violence of July 2009 as a “pot of boiling water” permanently on the verge of boiling over because the lid is screwed on too tight. Aynür, the language teacher told us “ethnic unity is like rubbing my warm cheek against their cold ass”. These are demands to be deemed worthy of the empathy and consideration which Tu Weiming insists China is deserves at the global level. However, these demands are silenced in an ongoing project to build the nation and project its model onto the world. The violence committed by Uyghurs on July 5th 2009 has been used by party-state to deny Uyghurs the right to empathy. The violent actions of the few clearly lacked empathy for ordinary Han who became ethnic targets. However, these actions have been represented by the party-state through essentialising discourses where Uyghurs become known to the world as “Islamic” “ethnic” “terrorists” in overlapping discourses of danger. These representations deny Uyghurs the right to empathy by producing an ethnicised cycle of violent representations.

It would be easy to conclude that history is merely repeating itself. Xinjiang is in China but not yet of it much as it was during the reign of the Qianlong emperor. The state has been made but the nation remains in a process of building and (re)building. China is a civilisation-state. However, this is not in the sense intended by Zhang Weiwei (2012) that China is unique because it is a unified state built on an unbroken singular civilisation. The CCP and nationalist intellectuals are projecting contemporary nationalist configurations of difference onto ancient boundaries in ways which produce ‘China’. The concept of civilisation is generally unworkable because there is too much conflict within what we think of as civilisations to be able to present them as singular and unified. Furthermore, it is hard to think of civilisation in the singular in China because there are multiple groups who speak languages from different language families and identify themselves as distinct groups. We can call ‘China’ a civilisation in the singular if we silence the histories of regions such as Xinjiang where multiple civilisations converged and

248 overlapped. To conceive of China as a civilisation projected onto contemporary spatial boundaries is to take the categories of Han and China as synonymous. This thesis showed that this is often taken to be the case in official and unofficial discourse on identity politics in contemporary China. However, it would seem both irrational and lacking in empathy to inscribe difference by claiming that China is a multi-ethnic state of 56 minzu and that these minzu are all of one civilisation defined through symbols of Han-ness. The party-state’s answer that identification with the nation demarcated through territorial borders comes above minzu is the standard response of all nationalists when facing competing conceptualisations within the national community.

Contemporary Chinese nationalism is perhaps best understood through William A Callahan’s argument that the imperial domains and sovereign territory work in “creative tension” to inscribe the PRC’s 21st century geobody on all of Chinese history (Callahan, 2009). The idea of “semantic hybridity” has been used throughout this thesis to frame how difference in China has never been transformed but is constantly reconfigured through contestation. China’s confrontation with European imperialism in the 19th century contributed to the reconfiguration of difference in China as nationalists sought to build a strong and modern sovereign state to resist the power of Europe and Japan. During the 19th century ‘civilisation’ became joined to imperialism in European thinking where the “white man’s burden” and la mission civilisatrice emerged to ‘civilise’ and or ‘dominate’ ‘inferior races’ (Cox, 2000, p.218). China’s confrontation with Europe has also contributed to the reconfiguration of difference where the ancient Confucian civilising mission of teaching barbarians to become civilised (jiaohua) has been modernised to teach barbarians they are part of an ancient Chinese civilisation and the contemporary Chinese state. To become a strong and modern state, China had to become more European in the sense of adopting international norms of sovereignty. However, this has also entailed European-isation by adapting 19th century racism to Confucianism and now to the construction of contemporary minzu policies. Stevan Harrell (1995) called China’s treatment of ethnicity a “civilising mission”. Today this traverses contemporary borders, for example, into where Chinese migrants believe they can develop their former imperial domain in what Pál Nyíri (2005) called the “yellow man’s burden”. What differentiates 19th century European Imperialism from contemporary Chinese nation-building in Xinjiang is that ‘China’ does not simply demand loyalty from ‘inferior races’, it demands identification with China and with the idea that Han are the superior ‘nucleus’ of the nation. Coercion and violence are key techniques in managing public discourse so that it does not challenge the party-state’s historical narrative or its minzu policies. Identity then becomes securitised so that alternatives can be repressed in the name of state-security. Minzu extinction (minzu xiaowang), discussed in chapter 2, is one of the stated end-goals of this civilising project. This extinction does not mean genocide. Extinction means conversion to China where the great revival of the Chinese people

249 entails a silencing of the darker, less harmonious side of China’s imperial past. However, this imperial past continues to haunt the lives of those who do not see themselves as Chinese in the present. They have the choice of identifying with China, staying silent, or being prosecuted for publicly articulating a different identity.

The conclusion here is not that “China is a civilisation pretending to be a state” as Lucian Pye famously claimed. Rather, China is a state pretending to be a civilisation. China is a sovereign state which seeks to civilise its frontiers not simply to make political and cultural boundaries congruent but to match imperial and modern spatial domains where imperial possessions such as Xinjiang are converted to the Chinese nation. This orders the nation into superior and inferior ethnic groups as different levels of civilisation conceptualised in singular and linear terms. China is being performed as a civilisation in ways which conceal the contestations over ethnic, national, and civilisational boundaries. For people in Xinjiang, multiplicity of contestations is the very essence of China because it is not minzu tuanjie which they see in their daily lives, it is separation between different minzu who live in the same physical space but occupy different civilisations.

250

Appendix 1: Lyrics to “One Family” Song – English Translation

One Family (English Translation)

Of all the red flowers154 which one are you, of all the willow green which one are you,

I don’t need to ask, you don’t need to say,

Red flowers and green willows are the colours of Spring.

Green mountain, which one are you, Water blossom, which one are you,

I don’t need to ask, you don’t need to say,

The picturesque scenes are joyous songs.

Home is one home, the country is great China,

If the family lives in harmony, all affairs will prosper155, it has you and it has me,

Home is one home, the country is great China,

We are all one family, you cannot separate you and me.

Warmth expels the cold winds, true love becomes a true heart,

We are all one family under the blue sky (),

We are all one heart, all one root,

Prosperity accompanies peace, we are all one family.

154 This has multiple meanings including a gift for a wedding, a bonus, or a tip, all of which imply prosperity. 155 家和万事兴 is from Confucius’ Analects. It means if the family lives in harmony, all affairs will prosper (家庭和睦就能兴旺).

251

Of all the red flowers which one are you, of all the willow green which one are you,

I don’t need to ask, you don’t need to say,

Red flowers and green willows are the colours of Spring.

Green mountain, which one are you, Water blossom, which one are you,

I don’t need to ask, you don’t need to say,

The picturesque scenes are joyous songs.

Help each other through trials and hardships, through danger we can see the truth,

Through sickness and wind we become indomitable, raging fire forges true gold,

You and I are one family, only love is this deep,

You and I are one family, our feelings are this genuine.

Warmth expels the cold winds, true love becomes a true heart,

We are all one family under the blue sky (tianxia),

We are all one heart, all one root,

Prosperity accompanies peace, we are all one family.

252

Appendix 2: Lyrics to “One Family” Song—Original Chinese

一家人

花红你是哪一朵，柳绿你是哪一棵，

不用我来问，不用你来说，

花红柳绿都是春色。

山青你是哪一座，水秀你是哪条河，

不用我来问，不用你来说，

山清水秀都是欢歌。

家是一个家，国是大中国，

家和万事兴，有你也有我，

家是一个家，国是大中国，

都是一家人，部分你和我。

温暖驱寒冷，真爱换真心，

同在蓝天下，都是一家人，

同是一颗心，同是一条根，

幸福伴样和，都是一家人。

花红你是哪一朵，柳绿你是哪一棵，

不用我来问，不用你来说，

花红柳绿都是春色。

253

山青你是哪一座，水秀你是哪条河，

不用我来问，不用你来说，

山青水秀都是欢歌。

相扶风雨中，危难见真情，

病风知劲草，烈火炼真金，

你我一家人，爱才那样深，

你我一家人，请才那样真。

温暖驱寒冷，真爱换真心，

同在蓝天下，都是一家人，

同是一颗心，同是一条根，

幸福伴样和，都是一家人。

254

BIBLIOGRAPHY

The 10 Conditions of Love (Ten Conditions) (2009). Film. Directed by Daniels, Jeff. : Umbrella Entertainment. A, Xingliang (1996) ‘Tujue Yuzu’ (‘Turkic Language Family’) in Yu, Taishan et al (eds) Xinjiang Gezu Lishi Wenhua Cidian (Dictionary of History and Culture of Every Xinjiang Minzu), Beijing: Zhonghua Shuju Chuban, p.316. Abramson, Kara (2012) ‘Gender, Uyghur Identity, and the Story of Nuzugum’, The Journal of Asian Studies, 71/3, pp.1-24. Abudoureyimu, Abulize (2006) ‘Jiaqiang Feiwuzhi Wenhua Yichan Baohu Gongzuo, Cujin Shehuizhuyi Hexie Xinjiang de Goujian’ (‘Strengthening the Protection of Intangible Cultural Heritage Work, Establishing a Mechanism for a Harmonious Xinjiang’) in Hao, Sumin (ed) Qiangjiu Baohu Feiwuzhi Wenhua Yichan Xibei ge Minzu zai Xingdong (Rescuing and Protecting Intangible Cultural Heritage: Every Northwestern Ethnic in Action), Beijing: Minzu Chubanshe, pp.65-69. Amit, Vered (2002a) 'Reconceptualizing Community' in Amit, Verid (ed) Realizing Community: Concepts, Social Relationships, and Sentiments, Routledge: London, pp.1-21. Amit, Vered (2002b) ‘An Anthropology Without Community?’ in Amit, Vered and Rapport, Nigel (eds) The Trouble With Community: Anthropological Reflections on Movement, Identity, and Collectivity, London: Pluto Press. Amnesty International (AI) (2003) “China: International community must oppose attempt to peaceful political activists as ‘terrorists’,”, December 19, 2003 [AI Index: ASA 17/040/2003]. Amnesty International (AI) (2010) ‘“Justice, Justice”: The July 2009 Protests in Xinjiang’ (Available at: http://www.amnesty.org/en/library/info/ASA17/027/2010/en), 2 July [Last Accessed: 16/06/12]. Amnesty International (AI) (2010b) ‘A Public Letter by Chinese Citizens Urging the Release of Uyghur Journalist Hailaite Niyazi’ (Available at: http://www2.amnesty.org.uk/blogs/countdown-china/public-letter-chinese-citizens- urging-release-uyghur-journalist-hailaite), 31 July [Last Accessed: 20/06/12]. Amnesty International (AI) (2012) ‘Nurmuhemmet Yasin’ (Available at: http://www2.amnesty.org.uk/groups/swansea/nurmuhemmet-yasin- %E5%8A%AA%E5%B0%94%E4%B9%B0%E4%B9%B0%E6%8F%90%E4%BA%9A%E 6%A3%AE) [Last Accessed: 01/09/12]. Anderson, Benedict (1991) Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso Press.

255

Anthony, Ross (2011) ‘Exceptionally Equal: Emergency States and the Production of Enemies in Xinjiang’, Inner Asia, 13/1, pp.51-72. Asia Weekly (Yazhou Zhoukan) (2009) ‘Caifang: Weizu NGO Gongzuozhe qian Xinjiang Fazhibao Zongbianbao Zhuren Hailaiteta zai Qiwu Shijian jiu Yujing’ (‘Interview: Heyrat, Uyghur NGO Worker and former Editor-in-Chief of the Xinjiang Legal News, gave Early Warnings Before 7-5’) (Available at: http://www.yzzk.com/cfm/Content_Archive.cfm?Channel=ag&Path=2311577102/30ag3a. cfm), 02/08/2009 [Last Accessed: 11/06/12]. Asian Development Bank (ADB) (2002) The 2020 Project: Policy Support in the PRC. Manila: Asian Development Bank. Baike (2012) ‘Kuerban Tulumu’ (Available at: http://baike.baidu.com/view/1903870.html?fromTaglist), [Last Accessed: 22/03/12]. Balzer, Harley (2004) ‘State and Society in Transitions from Communism' in Gries, Peter Hays and Rosen, Stanley (eds) State and Society in 21st-century China. New York: Routledge Curzon, pp.235-256. Barabantseva, Elena (2008) ‘From the Language of Class to the Rhetoric of Development: Discourse of ‘Nationality’ and ‘Ethnicity’ in China’, The Journal of Contemporary China, 17/56. Barabantseva, Elena (2009) ‘Development as Localization: Ethnic Minorities in China’s Official Discourse on the Western Development Project’, Critical Asian Studies, 41/2, pp.225-254. Barabantseva, Elena (2011) , Ethnic Minorities, and Nationalism: De-Centring China. Oxon: Routledge. Baranovitch, Nimrod (2003) China’s New Voices: Popular Music, Ethnicity, Gender, and Politics, 1978-1997. Berkeley: University of California Press. Barth, Frederick (1969) Ethnic Groups and Boundaries: The Social Organisation of Difference, Illinois: Waveland Press. BBC (2006) ‘Albania Takes Guantanamo Uyghurs’ (Available at: http://news.bbc.co.uk/1/hi/world/americas/4979466.stm), 6 May [Last Accessed: 01/09/12]. BBC (2009) ‘Scores Killed in China Protests’ (Available at: http://news.bbc.co.uk/1/hi/world/asia-pacific/8135203.stm), 6 July [Last Accessed: 11/06/12]. BBC (2009b) ‘Turkey Attacks China ‘Genocide’’ (Available at: http://news.bbc.co.uk/1/hi/world/asia-pacific/8145451.stm), 10 July [Last Accessed: 25/06/12]. Becquelin, Nicholas (2000) ‘Xinjiang in the Nineties’, The China Journal, 44, pp.65-90. Becquelin, Nicholas (2004a) ‘Staged Development in Xinjiang’, The China Quarterly, 178, pp.358-378.

256

Becquelin, Nicholas (2004b) ‘Criminalising Ethnicity: Political Repression in Xinjiang’, China Rights Forum, No.1 [Available at: http://www.hrichina.org/sites/default/files/oldsite/PDFs/CRF.1.2004/b1_Criminalizing1.20 04.pdf]. Beijing Cultural Heritage Protection Centre (2009): Available at: http://en.bjchp.org/english/indexen.asp [Last Accessed: 03/06/09]. Bellér-Hann, Ildikó (2001) 'Making the Oil Fragrant': Dealings With The Supernatural Among the Uyghurs in Xinjiang', Asian Ethnicity, 2/1, pp.9-23. Bellér-Hann, Ildikó (2002) ‘Temperamental Neighbours: Uighur-Han Relations in Xinjiang, ’ in Schlee, G (ed) Imagined Differences: Hatred and the Construction of Identity, New York: Palgrave. Bellér-Hann, Ildikó (2008) Community Matters in Xinjiang 1880-1949: Towards a Historical Anthropology of the Uyghur, Leiden: Brill. Bekri, Nur (2008) Nuer Baikeli zai Zizhiqu Ganbu Dahui shang de Jianghua (‘Nuer Baikeli’s Speech at the Autonomous Regional Government Congress’) (Available at: http://www.xj.xinhuanet.com/2008-09/11/content_14376020.htm) [last accessed: 31/05/12]. Bekri, Nur (2009a) ‘Nuer Baikeli Guanyu Wulumuqi “7-5” Shijian de Qingkuang Tongbao’ (‘Nur Bekri: Public Notification Regarding the Ürümchi 7-5 Incident’) in Xinjiang Uyghur Autonomous Region Party Commission Ministry of Information (eds) Jiaqiang Minzu Tuanjie, Weihu Xinjiang Wending: Xuanchuan Jiaoyu Cailiao (Yi) (Strengthening Ethnic Unity, Protecting Xinjiang Stability: Information Education Materials No.1), Wulumuqi: Xinjiang People’s Publishing Press. Bekri, Nur (2009b) ‘Nuer Baikeli zai Zizhiqu Tingju Ganbu Dahui Shang de Jianghua’ (‘Nur Bekri’s Speech at the Autonomous Regional Government’s Cadre Conference’) in Xinjiang Uyghur Autonomous Region Party Commission Ministry of Information (eds) Jiaqiang Minzu Tuanjie, Weihu Xinjiang Wending: Xuanchuan Jiaoyu Cailiao Yi (Strengthening Ethnic Unity, Protecting Xinjiang Stability: Information Education Materials No.1), Wulumuqi: Xinjiang People’s Publishing Press. Bense Shouyiren (2008) ‘Xianhua Han Minzu yu Zhonghua Minzu’ (‘A Complaint on the Han Ethnic Group and the Chinese Nation’) (Available at: http://www.uighurbiz.cn/socity/2008/1008/article_7277.html), UighurBiz, 8 October [Access no longer available]. Benson, Linda and Svanberg, Ingvar (1998) China’s Last Nomads: The History and Culture of China’s Kazakhs, New York: ME Sharpe. Berger, Mark T (2006) 'From Nation-Building to State-Building: The Geopolitics of Development, the Nation-State System, and the Changing Global Order', Third World Quarterly, 27 (1), p.5-25.

257

Berger, Stefan (ed) (2007) Writing the Nation: A Global Perspective. Basingstoke: Palgrave Macmillan. Bhabha, Homi K (2009) The Location of Culture. London: Routledge. Blank, Stephen (2003) ‘Xinjiang and China’s Security’, Global Economic Review, 32/4, pp.121- 148. Billig, Michael (1995) Banal Nationalism. London: Sage. Booth, Ken (ed) (2005) Critical Security Studies and World Politics. London: Lynne Rienner. Bourdieu, Pierre (1984) Distinctions: A Social Critique of the Judgment of Taste. Cambridge, MA: Harvard University Press. Bourdieu, Pierre (2008) Language and Symbolic Power. Cambridge: Polity Press. Bovingdon, Gardner (2002) ‘The Not-So Silent Majority: Uyghur Resistance to Han Rule in Xinjiang’, Modern China, 28 (1), pp.39-78. Bovingdon, Gardner (2004) ‘Autonomy in Xinjiang: Han Nationalist Imperatives and Uyghur Discontent’, East-West Center Policy Studies, 11: pp.ix-77. Bovingdon, Gardner (2004b) ‘Contested Histories’ in Starr, S.F (ed) Xinjiang: China’s Muslim Borderland, pp.353-374. New York: ME Sharpe. Bovingdon, Gardner (2010) The Uyghurs: Strangers in their Own Land. New York: Columbia University Press. Brophy, David (2011) Tending to Unite: The Origins of Uyghur Nationalism. Unpublished PhD thesis. Brubaker, Rogers (1998) 'Myths and Misconceptions in the Study of Nationalism', Hall, John (ed) The State of the Nation, Cambridge: Cambridge University Press. Brubaker, Rogers (2002) 'Ethnicity Without Groups', Archives Européenes de Sociologie, XLIII, 2, pp.163-189. Bulag, Uradyn E. (2002) The Mongols at China’s Edge: History and the Politics of National Unity, Oxford: Rowman and Littlefield. Burgess, Robert G (1982) 'The Unstructured Interview as Conversation' in Burgess, Robert G (ed) Field Research: A Sourcebook and Field Manual. London: Routledge, pp.107-110. Butler, Judith and Spivak, Gayatri Chakravorty (2007) Who Sings the Nation-State? Language, Politics, Belonging. Seagull Books: New York. Buzan, Barry et al (1998) Buzan, B et al (1998) Security: A New Framework for Analysis. London: Lynne Reinner. Callahan, William A (2004) Contingent States: and Transnational Relations, Minneapolis: University of Minnesota Press. Callahan, William A (2006) 'History, Identity, and Security: Producing and Consuming Nationalism in China', Critical Asian Studies, 38/2, pp.179-208. Callahan, William A (2007) ‘Trauma and Community: The Visual politics of Chinese Nationalism and Sino-Japanese Relations’, Theory and Event, 10/4.

258

Callahan, William A (2009) ‘The Cartography of National Humiliation and the Emergence of China’s Geobody’, Public Culture, 21/1, pp.141-173. Callahan, William A (2010) China: The Pessoptimist Nation, Oxford: Oxford University Press. Callahan, William A (2012) ‘Harmony, Unity, and Diversity in China’s World’, The Newsletter, 60, pp.22-23. Callahan, William A (2012b) ‘Who is Xi Jinping and Where Will He Lead China’ (Available at: http://www.opendemocracy.net/william-callahan/who-is-xi-jinping-and-where-will-he- lead-china), Open Democracy, 27 November [Last Accessed: 23/01/13]. Cameron, Angus and Palan, Ronen (1999) ‘The Imagined Economy: Mapping Transformations in the Contemporary State’, Milennium - Journal of International Studies, 28, pp.267-288. Campbell, David (1998) Writing Security: United States Foreign Policy and the Politics of Identity, Manchester: Manchester University Press. Campbell, David (2003) ‘Cultural Governance and Pictorial Resistance: Reflections on the Imaging of War’, Review of International Studies, 29, pp.57-73. Center for Strategic and International Studies (CSIS) (2007) Demography of HIV/AIDS in China, Washington: Center for Strategic and International Studies. CCTV (2009) ‘Xinjiang Artists Preserve Traditional Performances’, 01/06/09. CCTV (2009b) ‘Xinjiang is a Good Place’, 02/10/09 (Available at: http://english.cctv.com/program/worldwidewatch/20091002/101214.shtml) [Last Accessed: 23/03/12]. Cesaro, M Christina (2007) 'Polo, Laghman, So Say: Situating Uyghur Food Between Central Asia and China' in Bellér-Hann, I et al (2007) (eds) Situating the Uyghurs Between China and Central Asia, Aldershot: Ashgate. Chen, Kathy (1994) ‘Rags to Riches Story: How Rebiya Kader Made her Fortune’, Wall Street Journal, 21 September, A6. Chen-5 (2009) ‘Uncle Kurban Meets Chairman Mao and Rebiya Meets George Bush’ (Available at: http://bbs1.people.com.cn/postDetail.do?view=2&pageNo=1&treeView=0&id=93282208 &boardId=6), 17 July [Last Accessed: 20/06/12]. Chen Bao (2009a) ‘’Dixia Tongdao dou zhuang “Dianzi Yan”’ (‘Electric Eyes Installed in all Underpasses’), 22 September, p.A5. Chen Bao (2009b) ‘825 ming Xianfan bei Xingju’ (‘825 Criminal Suspects Detained’), 3 September, A3. Chen Bao (2009c) ‘Zizhiqu Tongbao Wushi Zhenci Shanghai Anjian Qingkuang’ (‘Regional Government Announces the Situation regarding the Cases of Syringe Assault Incidents’), 3 September, A3. Chen Bao (2009d) ‘Gongan Jiaqiang Shehui mian Fangkong, Gongjiaoche shang Cheche you Jingli’, 3 September, A3. Chen Bao (2009e) ‘Wushi Shehui Xunxu Dedao Youxiao Kongzhi’, 5 September, A1.

259

Chen Bao (2009f) ‘7-5 Zhuananzu yi Pibu 237 Ren’, 12 September. Chen Bao (2009g) ‘Quanjiang Zhuahuo 75 ming ‘Zhenci’ Zuifan’, 16 September, A1. Chen Bao (2009h) ‘Wei Jingwai Daoqu, Cishen, Shoumai, Feifa Tigong Guojia Mimi, Qingbaozui’, 15 September. Chen Bao (2009i) ‘Zhuanjia Jiedu “Guanyu Yifa Daji Zhenci Weifa Fanzui Huodong de Gonggao”’, 9 September. Chen Bao (2009j) ‘Zhongxiaoxue SheliWeiwen Anquanyuan’, 3 September. Chen Bao (2009k) ‘Dianli Xidezuo, Minjian Hu Guoqi’, 5 September. Chen Bao (2009L) ‘Zhencian Cong Zhongliangxing bing wu Budang’, 18 September, A4. Chen Bao (2009m) ‘Zui Ming Xiangtong Weihe Zhongxing bu Tong’, 18 September, A5. Chen Bao (2009n) ‘Xi’an Fasheng Yisi Zhenci Shanghaian’, 17 September. Chen Bao (2009o) ‘Xi’an Po Liang qi Zhenza’an’, 22 September, A10. Chen Bao (2009p) ‘Meng Jianzhu lai Wu Jiancha Zhidao Weiwen Gongzuo’ (‘Meng Jianzhu Inspects Urumqi to Lead Stability Work’), 5 September, A1. Chen Bao (2009q) ‘Haizemen Youge Weiwuerzu Ma’ (‘Children Have a Uyghur Mother’), 15 December, A4. Chen Bao (2009r) ‘Tongxu weile Wo Hunli Gai Riqi’ (‘My Classmate Changed his Wedding Date for Me’), 15 December, A4. Chen Bao (2009s) ‘Rouzi jie Fangjia Tonghi’ (‘Roza Holiday Notification’), Chen Bao, 15 September, A2. Chen Bao (2009t) ‘Huandu Rouzi Jie’ (‘Warmly Celebrate Roza’), Chen Bao, 22 September, Cover. Chan, Koonchung (2010) Shengshi: Zhongguo 2013 (The Fat Years: China 2013). Hong Kong: Oxford University Press. Chen, Yangbin (2008) Muslim Uyghur Students in a Chinese Boarding School: Social Recapitalisation as a Response to Ethnic Integration, Plymouth: Lexington. Child, Ben (2009) ‘Old Suspicions Magnified Mistrust into Ethnic Riots in Urumqi’ (Available at: http://www.guardian.co.uk/film/2009/sep/22/chinese-protesters-hack-website), The Guardian, 22 September [Last Accessed: 14/09/12]. China Daily (2009) ‘Xinjiang Emerging from Riot Shadow’, 2 September. China Digital Times (2009) ‘Death Toll in Xinjiang Rises to 184 (Updated)’ (Available at: http://chinadigitaltimes.net/2009/07/death-toll-in-xinjiang-violence-rises-to-184/), 10 July [Last Accessed: 11/06/12]. China International Travel Service (CITS): http://www.bestourchina.com/citsonlineWeb/online/homepage/index.htm [Last Accessed: 03/06/09]. China Islamic Association (2006) Musilin Aiguozhuyi Jiaocheng (‘Patriotic Muslim Coursebook’), Beijing: Zongjiao Wenhua Chubanshe.

260

China Labor Watch (2009) ‘Labor Violations Exacerbate Ethnic Tensions in South China’ (Available at: http://www.chinalaborwatch.org/20090706uyghur.htm#), 7 July [Last Accessed: 11/06/12]. China Radio International Online International News (2005) ‘Sixiang zhengzhi jiaoyu shi neidi xinjiang gaozhong ban de shouyao renwu’, (‘Political Thought Education is the Primary Function of Xinjiang Classes’) 24/6/2005 [available at: http://gb.cri.cn/3821/2005/06/24/[email protected]] (Last Accessed: 21/2/09). Chow, Kai-Wing (2001) 'Narrating the Nation, Race, and National Culture: Imagining the Hanzu Identity in Modern China' in Chow, Kai-Wing et al (eds) Constructing Nationhood in Modern East Asia, Ann Arbor: University of Michigan Press, pp.47-85. Chowdry, Geeta (2007) ‘Edward Said and Contrapuntal Reading: Implications for Critical Interventions in International Relations’, Millennium- Journal of International Studies, 36/1, pp.101-116. Clarke, Michael E. (2007a) "The Problematic Progress of 'Integration' in the Chinese State's Approach to Xinjiang, 1759-2005", Asian Ethnicity, 8/3, pp.261-289. Clarke, Michael E. (2007b) 'China's Internal Security Dilemma and the "Great Western Development": The Dynamics of Integration, Ethnic Nationalism, and Terrorism in Xinjiang, Asian Studies Review, pp.323-342. Clarke, Michael E. (2010) Xinjiang and China’s Rise in Central Asia. London: Routledge. CNN (2007) ‘Chinese Architect Slams Olympics Pretend Smile’ (Available at: http://edition.cnn.com/2007/WORLD/asiapcf/08/13/china.olympics.reut/), 13 August [Last Accessed: 01/09/12]. Cohen, Anthony P (1985) The Symbolic Construction of Community. Routledge: London. Cohen, Anthony (1996) 'Personal Nationalism: A Scottish View of Some Rites, Rights, and Wrongs', American Ethnologist, 23/4, pp.802-815. Cohen, Paul (2010) Discovering History in China: American Historical Writing on the Recent Chinese Past, New York: Columbia University Press. Congressional Executive Commission on China (CECC) (2005) ‘Uighur Author Sentenced to Prison for “Inciting Splittism”’ (Available at: http://www.cecc.gov/pages/virtualAcad/index.phpd?showsingle=6759), 10 February [Last Accessed: 01/09/12]. Connolly, William E (2002) Identity/Difference: Democratic Negotiations of Political Paradox. Minneapolis: University of Minnesota Press. Connor, Walker (1984) The National Question in Marxist-Leninist Theory and Strategy, Princeton: Princeton University Press. Constitution of the People’s Republic of China (2004) Beijing: Foreign Languages Press. Cox, Robert (1981) ‘Social Forces, States, and World Orders: Beyond International Relations Theory’, Milennium: Journal of International Studies, 10/2, pp.126-155.

261

Cox, Robert (2000) ‘Thinking About Civilizations’, Review of International Studies, 26, pp.217- 234. Da, Wei (2010) Shei shi Zhongguo Ren? (Who is Chinese?). Changsha: Hunan People’s Publishing Press. Dankoff, Robert (1983) ‘Introduction’ in Dankoff, Robert (ed) Hajib, Yusuf-Khass (1983) Wisdom of Royal Glory: A Turko-Islamic Mirror For Princes, Chicago: Chicago University Press. Dautcher, Jay (2009) Down a Narrow Road: Identity and Masculinity in a Uyghur Community in Xinjiang China. Cambridge, Massachusetts: Harvard University Press. Davies, Norman (1997) Europe: A History. Pimlico: London. Dawut, Rahila (2007) 'Shrine Pilgrimage and Sustainable Tourism among the Uyghurs: Central Asian Ritual Traditions in the Context of China's Development Policies' in Bellér-Hann, I et al (2007) (eds) Situating the Uyghurs Between China and Central Asia, Aldershot: Ashgate. Dillon, Michael (2002) Xinjiang: China’s Muslim Far North West. London: Routledge Curzon. Dikötter, Frank (1994) ‘Racial Identities in China: Context and Meaning’, The China Quarterly, 138, pp.404-412. Doty, Roxanne (2000) ‘Desire All the Way Down’, Review of International Studies, 26, pp.137- 139. Duara, Prasenjit (1988) Culture, Power, and the State, Stanford: Stanford University Press. Duara, Prasenjit (1995) Rescuing History from the Nation: Questioning Narratives of Modern China. London: University of Chicago Press. Duara, Prasenjit (1998) ‘The Regime of Authenticity: Timelessness, Gender, and National History in Modern China’, History and Theory, 37/3, pp.287-308. Duara, Prasenjit (2009) The Global and Regional in China’s Nation-Formation. Routledge: London. Dwyer, Arienne M (2005) ‘The : Uyghur Identity, Language, Policy, and Political Discourse’, East-West Center Policy Studies, 15. East Turkestan Information Center [Available at: http://www.uygur.org/] (last accessed: 09/03/09). Edwards, Louise (2000) ‘Policing the Modern Woman in Republican China’, Modern China, 26/2, pp.115-147. Elliot, Mark (2001) The Manchu Way: the Eight Banners and Ethnic Identity in Late Imperial China. Stanford: Stanford University Press. Elliot, Mark (2011) ‘Hushuo: The Northern Other and Han Ethnogenesis’ in Mullaney, Thomas S et al (eds) Critical Han Studies: the History, Representation, and Identity of China’s Majority. Berkeley: University of California Press.

262

Elias, Juanita (2005) ‘Stitching Up the Labour Market’, International Feminist Journal of Politics, 7/1, pp.90-111. Engels, Friedrich (2010) The Origin of the Family, Private Property, and the State. London: Penguin Books. Ethnic Affairs Commission (2009) Xinjiang Wenhua Zhishi Duben (Xinjiang Cultural Knowledge Study Guide), Beijing: People’s Publishing Press. Ethnic Unity Education Board (2009) Minzu Lilun Changshi (Common Knowledge of Ethnic Theory). Beijing: Central Television and Broadcasting Publishing House. Evans, Harriet (2008) ‘Gender in Modern Chinese Culture’, in Louie, Kam (ed) The Cambridge Companion to Modern Chinese Culture, Cambridge: Cambridge University Press, p.68-90. Fairbank, John (ed) (1968) The Chinese World Order, Cambridge, MA: Harvard University Press. Fei, Xiaotong (1980) ‘Guanyu Woguo Minzu Shibie Wenti’ (‘On China’s Ethnic Classification Project’), Zhongguo Shehui Kexue, 1. Fei, Xiaotong (1988) ‘Plurality and Unity in the Configuration of the Chinese People’, Tanner Lectures on Human Values, 15th & 17th November (available at: http://www.tannerlectures.utah.edu/lectures/documents/fei90.pdf) Feng, Jianyong and Aney, Adil (2009) ‘Build a Bridge that Binds Workers of all Hues’, China Daily, 3 September. Financial Times (FT) (2010) ‘Development Unlikely to Drown Out Disharmony’ (Available at: http://www.ft.com/cms/s/0/f41cd062-85fa-11df-bc22- 00144feabdc0.html#axzz22rBf84wk), 2 July [Last Accessed: 07/08/12]. Fletcher, Joseph (1978) ‘Ch’ing Inner Asia c.1800’ in Fairbank, John (ed) The Cambridge History of China, Vol.10, Late Ch’ing, 1800-1911, pt.1, Cambridge: Cambridge University Press. Forbes, Andrew (1986) and Muslims in Chinese Central Asia. Cambridge: Cambridge University Press. Foreign Correspondent’s Club of China (FCCC) (2009a) ‘TV Crew Detained in Urumqi’ (Available at: ‘http://www.fccchina.org/2009/07/06/tv-crew-detained-in-urumqi/’), 6 July [Last Accessed: 22/06/12]. Foreign Correspondent’s Club of China (FCCC) (2009b) ‘Reporter in Urumqi Shoved into Van’ (Available at: http://www.fccchina.org/2009/07/07/reporter-in-urumqi-shoved-into-van/), 7 July [Last Accessed: 22/06/12]. Foreign Correspondent’s Club of China (FCCC) (2009c) ‘China Must Stop Harassing Reporters in Xinjiang’ (Available at: http://www.fccchina.org/2009/07/07/china-must-stop-harassing- reporters-in-xinjiang/), 7 July [Last Accessed: 22/06/12]. Foreign Correspondent’s Club of China (FCCC) (2009d) ‘China Should Allow Reporters Free Movement in Xinjiang’ (Available at: http://www.fccchina.org/2009/07/11/china-should- allow-reporters-free-movement-in-xinjiang/), 11 July [Last Accessed: 22/06/12].

263

Foreign Correspondent’s Club of China (FCCC) (2009e) ‘Open Letter on Reporting Conditions in Xinjiang’ 20 July (Available at: http://www.fccchina.org/2009/07/20/open-letter-on- reporting-conditions-in-xinjiang/), [Last Accessed: 22/06/12]. Foreign Correspondent’s Club of China (FCCC) (2009f) ‘Urumqi Police Seize Cameras During Protest’ (Available at: http://www.fccchina.org/2009/09/04/urumqi-police-seize-cameras- during-protest/), 4 September [Last Accessed: 22/06/12]. Foreign Correspondent’s Club of China (FCCC) (2009g) ‘Urumqi Armed Police Beat Journalists, Confiscate Tapes’ (Available at: http://www.fccchina.org/2009/09/04/urumqi-armed- police-beat-journalists-confiscate-tapes/), 4 September [Last Accessed: 22/06/12]. Foreign Correspondent’s Club of China (FCCC) (2009h) ‘Urumqi Paramilitary Beat Journalists, Damage Gear’ (Available at: http://www.fccchina.org/2009/09/05/urumqi-paramilitary- beat-journalists-damage-gear/), 5 September [Last Accessed: 22/06/12]. Foreign Correspondent’s Club of China (FCCC) (2009i) ‘FCCC Condemns Beating of Journalists in Urumqi’ (Available at: http://www.fccchina.org/2009/09/05/fccc-condemns-beating-of- journalists-in-urumqi/), 5 September [Last Accessed: 22/06/12]. Foster, Peter (2009) ‘300 Uighurs Stage Fresh Protest in Urumqi’ (Available at: http://www.telegraph.co.uk/news/worldnews/asia/china/5764906/China-riots-300- Uighurs-stage-fresh-protest-in-Urumqi.html), The Telegraph, 7 July [Last Accessed: 14/09/12]. Foucault, Michel (1991/1976) The Will to Knowledge: The History of Sexuality Volume 1, London: Penguin Books. Fravel, Taylor (2005) ‘Regime Insecurity and International Cooperation: Explaining China's Compromises in Territorial Disputes’, International Security, 30/2, 2005. Fravel, Taylor (2007) ‘Power Shifts and Escalation: Explaining China's Use of Force in Territorial Disputes’, International Security, 32/3, pp. 44-83. Freeman, Charles W. & Wen Jin Yuan (2012) ‘The Influence and Illusion of China’s New Left’, The Washington Quarterly, 35/1, pp.65-82. Friederich, Michael (2007) ‘Uyghur Literary Representations of Xinjiang Realities’ in Bellér-Hann, I et al (2007) (eds) Situating the Uyghurs Between China and Central Asia, pp. 89-108. Aldershot: Ashgate. Fukuyama, Francis (1989) ‘The End of History’, National Interest, Summer 1989. Fukuyama, Francis (2004a) State-Building: Governance and World Order in the 21st Century, New York: Cornell University Press. Fukuyama, Francis (2004b) 'Nation-Building 101' in The Atlantic Monthly, 293 (1), p.159-162. Fuller, Graham and Lipman, Jonathan (2004) ‘Islam in Xinjiang’, in Starr, S.F (ed) Xinjiang: China’s Muslim Borderland, pp.320-352. New York: ME Sharpe. Gaubatz, Piper Rae (1996) Beyond the Great Wall: Urban Form and Transformation on the Chinese Frontier. Stanford University Press: Stanford.

264

Geertz, Clifford (1973) The Interpretation of Cultures: Selected Essays by Clifford Geertz. New York: Basic Books. Gellner, Ernest (1996) ‘The Nation: Real or Imagined?’, Nations and Nationalism, 2/3, pp.357- 370. Gellner, Ernest (2006) Nations and Nationalism. Oxford: Wiley-Blackwell. Gladney, Dru (1990) ‘The Ethnogenesis of the Uighur’, Central Asian Survey, 9/1, pp.1-28. Gladney, Dru (1996) Muslim Chinese: Ethnic Nationalism in the People’s Republic. Massachusetts: Harvard University Press. Gladney, Dru (2004) Dislocating China: Muslims, Minorities, and Other Subaltern Subjects, London: Hirst & Co. Global Times (2009) ‘Xinjiang Migrant Labour-Plan Combats Poverty’ (Available at: http://china.globaltimes.cn/chinanews/2009-07/448670.html), 20 July [Last Accessed: 11/06/12]. Gold, Thomas B (1989) 'Guerilla Interviewing Among the Getihu' in Link, Perry et al (eds) Unofficial China: Popular Culture and Thought in the People's Republic, Boulder: Westview Press, pp.175-192. Goldstein, Avery (2001) 'The Diplomatic Face of China's Grand Strategy: A Rising Power's Emerging Choice', The China Quarterly, 168, pp.835-864. Gong, Yufeng (2009) ‘Xinjiang: A Fairyland Ever and Forever’ (Available at: http://eg.china- embassy.org/eng/rdwt/gert2444/t575748.htm), 28 July [Last Accessed: 14/01/13]. Goodman, David (2002) ‘The Politics of the West: Equality, Nation-Building and Colonisation’, Provincial China, 7(2): pp.127-150. Goodman, David (2008) ‘Why China Has No New Middle Class: Cadres, Managers, and Entrepeneurs’ in Goodman, David (ed) The New Rich in China: Future Rulers, Present Lives. Oxon: Routledge, pp.23-38. Goodman, D (2004a) ‘The Campaign to “Open Up the West”: National, Provincial-Level, and Local Perspectives’, The China Quarterly, 178: pp.317-334. Gramsci, Antonio (1935) Selections from Cultural Writings in Forgacs, David (ed) (1999) The Antonio Gramsci Reader: Selected Writings 1916-1935. London: Lawrence and Wishart. Gries, Peter Hays (2004) ‘Popular Nationalism and State Legitimation in China’ in Gries, Peter Hays and Rosen, Stanley (eds) State and Society in 21st Century China: Crisis, Contention, and Legitimation, London: Routledge, pp.180-194. Griffin, Penny (2007) ‘Refashioning IPE: What and How Gender Analysis Teaches International (Global) Political Economy’, Review of International Political Economy, 14/4, pp.719-736. Guangming Wang News Centre (2005) ‘Xinjiang pojie “shuangyu” jiaoyu nanti’ (‘Xinjiang Resolves Bilingual Education Problem’), 19/12/2005 [available at: http://www.gmw.cn/01gmrb/2005-12/19/content_347901.htm] (Last Accessed: 1/2/2009).

265

Hall, Stuart (1996) “New Ethnicities,” in Morley, David and Chen Kuan-Hsing (eds) Stuart Hall: Critical Dialogues in Cultural Studies. London: Routledge. Hajib, Yusuf-Khass (1983) Wisdom of Royal Glory: A Turko-Islamic Mirror For Princes, Chicago: Chicago University Press. Harrell, Stevan (1990) 'Ethnicity, Local Interests, and the State', Comparative Studies in Society and History, 32/3, p.515-548. Harrell, Stevan (1995) ‘Introduction: Civilizing Projects and the Reaction to them’, in Harrell, Stevan (ed) Cultural Encounters on China’s Frontiers, Washington: University of Washington Press. Harris, Rachel (2001) ‘Cassettes, Bazaars and Saving the Nation: The Uyghur Music Industry in Xinjiang, China’ in Craig, Tim and King, Richard (eds) Global Goes Local: Popular Culture in Asia. Vancouver: University of British Columbia Press, pp.265-283. Harris, Rachel (2005) ‘Wang Luobin: Folk Song King of the Northwest or Song Thief? Copyright, Representation, and Folk Songs’, Modern China, 31, pp.381-408. Harris, Rachel (2008) The Making of a Musical Canon on Chinese Central Asia: The Uygur Twelve Muqam, Ashgate: Aldershot. Harris, Rachel (2011) ‘Invitation to a Mourning Ceremony: Perspectives on the Uyghur Internet’, Inner Asia, 13/1, pp.27-49. Harrison, Henrietta (2000) The Making of the Republican Citizen: Political Ceremonies and Symbols in China, 1911-1929, Oxford: Oxford University Press. He, Baogong (2006) ‘ with Chinese Characteristics’ in Kymlicka, Will and He, Baogong (eds) Multiculturalism in Asia. Oxford: Oxford University Press. Heaphy, Brian (2011) ‘Gay Identities and the Culture of Class’, Sexualities, 14/42, pp.42-62. Heathershaw, John and Megoran, Nick (2011) ‘Contesting Danger: A New Agenda for Policy and Scholarship on Central Asia’, International Affairs, 87/3, pp.589-612. Hechter, Michael (1975) Internal Colonialism: The Celtic Fringe in British National Development, New Jersey: Transaction. Hershatter, Gail (1993) 'The Subaltern Talks Back: Reflections on Subaltern Theory and Chinese History', Positions, 1/1, p.103-130. Hershatter, Gail (2007) ’s Long Twentieth Century. Berkeley: University of California Press. Hess, Steve (2009) ‘Dividing and Conquering the Shop Floor: Uyghur Labour Export and Labour Segmentation in China’s Industrial East’, Central Asian Survey, 28/4, pp.403-416. Hobsbawm, Eric J (1990) Nations and Nationalism Since 1780: Programme, Myth, Reality. Cambridge: Cambridge University Press. Holbig, Heike (2004) ‘The Emergence of the Campaign to Open Up the West: Ideological Formation, Central Decision-Making and the Role of the Provinces’, The China Quarterly, 178: pp.335-357.

266

Hollis, Martin and Smith, Steve (1991) ‘Beware of Gurus: Structure and Action in International Relations’, Review of International Studies, 17, pp.393-410. Hu, Angang and Hu Lianhe (2012) ‘Di Er Dai Minzu Zhengce: Cujin Minzu Jiaorong Yiti he Fanhua Yiti’ (‘The Second Generation of Minzu Policies: Promoting Minzu Fusion and Prosperity in an Organic Whole’) (Available at: http://www.mzb.com.cn/html/Home/report/293093-1.htm), 10 April [Last Accessed: 24/08/12]. Hu, Angang (2012b) Zhongguo 2020: Yige Xinxing Chaoji Daguo (‘China in 2020: A New Type of Superpower’), : Zhejiang Renmin Chubanshe. Hu, Jintao (2005) ‘Written Speech by H.E. Hu Jintao President of the People's Republic of China At the High-level Plenary Meeting of the ' 60th Session’ (Available at: http://www.fmprc.gov.cn/eng/wjdt/zyjh/t213091.htm), Ministry of Foreign Affairs of the People’s Republic of China,16 September [Last Accessed: 25/06/12]. Hu, Jintao (2007) ‘Hold High the Great Banner of Socialism With Chinese Characteristics and Strive For New Victories In Building a Moderately Prosperous Society in all Respects: Report to the Seventeenth National Congress of the Communist Party of China on October 15, 2007’ in Documents of the 17th National Congress of the Communist Party of China, Beijing: Foreign Languages Press, pp.1-73. Hu, Jintao (2009) ‘Hu Jintao Zhute Zhongyang Zhengzhiju Changwu Weiyuan Huiyi Yanjiu Buzhi Weihu Xinjiang Shehui Wending Gongzuo’ in Xinjiang Uyghur Autonomous Region Party Commission Ministry of Information (eds) Jiaqiang Minzu Tuanjie, Weihu Xinjiang Wending: Xuanchuan Jiaoyu Cailiao Yi (Strengthening Ethnic Unity, Protecting Xinjiang Stability: Information Education Materials No.1), Wulumuqi: Xinjiang People’s Publishing Press. Human Rights Watch (HRW) (2009) ‘“We Are Afraid to Even Look for Them”: Enforced Disappearances in the Wake of Xinjiang’s Protests’ (Available at: http://www.hrw.org/sites/default/files/reports/xinjiang1009webwcover.pdf), 1 October [Last Accessed: 16/06/12]. Human Rights Watch (HRW) (2009b) ‘China: Forcibly Returned Uighur Asylum Seekers at Risk’ (Available at: http://www.hrw.org/news/2009/12/22/china-forcibly-returned-uighur- asylum-seekers-risk), 22 December [Last Accessed: 20/06/12]. Human rights Watch (HRW) (2011) ‘China: Account for Forcibly Returned Uighurs’ (Available at: http://www.hrw.org/news/2011/09/02/china-account-forcibly-returned-uighurs), 2 September [Last Accessed: 20/06/12]. Hunan Dianshitai (2009) Baqian Xiang Shangshan (‘8,000 Flowers Go Up the Mountain’), DVD, Hunan Dianshitai: Hunan. Hume, David (1739/2003) A Treatise of Human Nature. Dover Publications: New York.

267

Huntington, Samuel P (1996) The Clash of Civilizations and the Remaking of World Order. Touchstone: New York. Index on Censorship (2008) ‘Truth to Power: An Interview with Ai Weiwei’ (Available at: http://www.indexoncensorship.org/wp-content/uploads/2008/07/ai-weiwei_interview.pdf), 11/06/2008 [Last Accessed: 04/04/12]. Inside-Out China (2012) ‘On Bo Xilai’s Chongqing Model’ (Available at: http://insideoutchina.blogspot.co.uk/2012/03/bo-xilais-chongqing-model.html), 20 March [Last Accessed: 24/08/12]. Jacobs, Justin (2008) ‘How Turkestan Became Chinese: Visualising Zhang Zhizong's Tian Shan Pictorial and Xinjiang Youth Song and Dance Troupe', The Journal Of Asian Studies, 47/2, pp.545-591. Jeffries, Elaine (2008) ‘Advanced Producers or Moral Polluters? China’s Bureaucrat- Entrepreneurs and Sexual Corruption’ in Goodman, David (ed) The New Rich in China: Future Rulers, Present Lives. Oxon: Routledge, p.229-244. Jia Jianfei (2011) ‘Whose Xinjiang? The Transition in Chinese Intellectuals’ Imagination of the “New Dominion” During the Qing Dynasty’, Harvard-Yenching Institute Working Paper Series. Hu, Wei and Shao, Dazhen (2012) Xinjiang Youhua Quanguoxing Qingxi Shenzhou / The Affection for Motherland. Jilin: Jilin Meishu Chubanshe. Johnston, Alastair Iain (1995) Cultural Realism: Strategic Culture and Grand Strategy in Chinese History. Princeton, New Jersey: Princeton University Press. Johnston, Alastair Iain (2003) ‘Is China a Status Quo Power?’, International Security, vol. 27, no. 4, Spring 2003, pp. 5-56. Kadeer, Rebiya (2009) Dragon Fighter: One Woman’s Epic Struggle for Peace with China. Carlsbad, CA: Kales Press. Kadeer, Rebiya (2012a) ‘Beijing Stirs Ethnic Hatred in Xinjiang’ (Available at: http://online.wsj.com/article/SB10001424052702304211804577502141575598740.html), The Wall Street Journal, 2 July [Last Accessed: 17/09/12]. Kadeer, Rebiya (2012b) ‘Testimony of Ms. , Uyghur Democracy Leader’ (Available at: http://foreignaffairs.house.gov/112/HHRG-112-FA00-WState-KadeerR-20120725.pdf), House Committee on Foreign Affairs, Investigating the Chinese Threat, Part Two: Human Rights Abuses, Torture, and Disappearances, 25 July [Last Accessed: 17/09/12]. Kamalov, A (2007) ‘The Uyghurs as Part of Central Asian Commonality: Soviet Historiography on the Uyghurs’ in Bellér-Hann, I et al (2007) (eds) Situating the Uyghurs Between China and Central Asia, Aldershot: Ashgate. Kerr, David (2011) ‘Paradoxes of Tradition and Modernity at the New Frontier: China, Islam, and the Problem of “Different Heavens”’ in Callahan, William A and Barabantseva, Elena

268

(eds) China Orders the World: Normative Soft Power and Foreign Policy. Woodrow Wilson Center Press: Washington DC, pp.143-179. Khalid, Adeeb (2007) Islam After Communism, Berkeley: University of California Press. Kipnis, Andrew (2006) ‘Suzhi: A Keyword Approach’, The China Quarterly, 186, pp.295-313. Kurlantzick, Joshua (2007) China’s Charm Offensive: How China’s Soft Power is Transforming the World. New York: Caravan Books. Lai, Harry Hongyi (2002a) ‘China’s Western Development Program: Its Rationale, Implementation, and Prospects’, Modern China, 28(4): pp.432-466. Lam, Willy (2009) ‘The Xinjiang Crisis: A Test for Beijing’s Carrot-and-Stick Strategy’, China Brief, 9/15 (Available at: http://www.jamestown.org/programs/chinabrief/single/?tx_ttnews%5Btt_news%5D=353 07&tx_ttnews%5BbackPid%5D=459&no_cache=1) [Last Accessed: 06/12/11]. Laogai Research Foundation (2008) Laogai Handbook, Washington DC. Leibold, James (2007) Reconfiguring Chinese Nationalism: How the Qing Frontier and its Indigenes Became Chinese, New York: Palgrave. Leibold, James (2010) The Beijing Olympics and China's Conflicted National Form, The China Journal, 63, pp.1-24. Leibold, James (2012a) ‘Can China Have a Melting Pot?’ (Available at: http://thediplomat.com/china-power/can-china-have-a-melting-pot/), The Diplomat, 23 May [Last Accessed: 11/07/12]. Leibold, James (2012b) ‘Toward a Second Generation of Ethnic Policies’ (Available at: http://www.jamestown.org/uploads/media/cb_07_02.pdf), China Brief, 12/13, 6 July [Last Accessed: 11/07/12]. Lewis, Bernard (1990) ‘The Roots of Muslim Rage’ (available at: http://www.theatlantic.com/magazine/archive/1990/09/the-roots-of-muslim- rage/304643/), The Atlantic, September [Last Accessed: 15/01/13]. Li, Dezhu (2007) Guojia Minwei Zhuren Li Dezhu Zai Quanguo Minwei Zhuren Huiyishang de Jianghua (National Commission for Ethnic Affairs Chairman Li Dezhu’s Speech at the 2007 National Commission for Ethnic Affairs Conference) (Available at: http://news.sohu.com/20071229/n254377534.shtml) [Last Accessed: 03/06/09]. Li, Sheng (2006) Dong Tujuesitan” Fenliezhuyi de Youlai yu Fazhan (The Origins and Development of “East Turkestan” Separatism), Urumchi: Xinjiang People’s Press. Ma, Rong (2007) 'A New Perspective in Guiding Ethnic Relations in the 21st Century: 'Depoliticization' of Ethnicity in China', The University of Nottingham China Policy Institute, Discussion Paper 21. Ma, Rong (2009) 'The Development of Minority Education and the Practice of Bilingual Education in Xinjiang Uyghur Autonomous Region', Frontier Education China, 4/2, pp.188-251.

269

Ma, Rong (2010) ‘Hanhua hai shi Xiandaihua’ (‘Hanification is still Modernisation’) (Available at: http://www.21ccom.net/articles/zgyj/ggzhc/article_2010121726544.html), Gongshi Wang, 17 December [Last Accessed: 13//07/12]. Mackerras, Colin (2009) 'Xinjiang and Central Asia Since 1990: Views From Beijing and Washington and Sino-American Relations' in Mackerras, Colin and Clarke, Michael (eds) China, Xinjiang, and Central Asia: History, Transition, and Crossborder Interaction into the 21st Century, London: Routledge, p.133-150. Mao, Zedong (1923) ‘The Cigarette Tax’ (Available at: http://www.marxists.org/reference/archive/mao/selected-works/volume- 6/mswv6_08.htm) [Last Accessed: 29/05/12]. Mao, Zedong (1926) ‘Analysis of the Classes in Chinese Society’ (Available at: http://www.marxists.org/reference/archive/mao/selected-works/volume-1/mswv1_1.htm) [Last Accessed: 29/05/12]. Mao, Zedong (1937) ‘On Contradiction’ (Available at: http://www.marxists.org/reference/archive/mao/selected-works/volume- 1/mswv1_17.htm) [Last Accessed: 29/05/12]. Marx, Karl (1869/1963) The Eighteenth Brumaire of Louis Bonaparte. New York: International Publishers. McMillen, Donald H. (1979) Chinese Communist Power and Policy in Xinjiang, 1949-1977. Boulder, CO: Westview Press. Meng, Yue (1993) ‘Female Images and ’ in Barlow, Tani F (ed) Gender Politics in Contemporary China: Writing and Feminism. Durham, NC: Duke University Press. Millward, James A (1994) ‘A Uyghur Muslim in Qianlong’s Court: The Meaning of the Fragrant Concubine’, The Journal of Asian Studies, 53/2, pp.427-458. Millward, James A (1996) ‘New Perspectives on the Qing Frontier’ in Hershatter, Gail et al (eds) Remapping China: Fissures in Historical Terrain, Stanford: Stanford University Press, pp.113-129. Millward, James A (2004) ‘Violent Separatism in Xinjiang: A Critical Assessment’, Policy Studies 6, East-West Center, Washington. Millward, James A and Perdue, Peter (2004) ‘Political and Cultural History of the Xinjiang Region Through the Late 19th Century’ in Starr, S.F (ed) Xinjiang: China’s Muslim Borderland, pp.27-62. New York: ME Sharpe. Millward, James (2005) ‘Uyghur Art Music and the Ambiguities of Chinese Silk Roadism in Xinjiang’, Silk Road, 3 (1) [available at: http://www.silk- road.com/newsletter/vol3num1/3_uyghur.php] (Last Accessed: 13/04/09). Millward, James A (2007) Eurasian Crossroads: A History of Xinjiang. London: Hurst and Company.

270

Millward, James A (2009) 'Positioning Xinjiang in Eurasian and Chinese History: Differing Visions of the 'Silk Road'' in Mackerras, Colin and Clarke, Michael (eds) China, Xinjiang, and Central Asia: History, Transition, and Crossborder Interaction into the 21st Century, London: Routledge, p.55-74. Millward, James A (2009b) ‘Does the Urumchi Violence Mark a Turning Point?’, Central Asian Survey, 28/4, pp.347-360. Millward, James A (2011) ‘Being Blacklisted by China and What Can Be Learned From It’ [Available at: http://www.thechinabeat.org/?p=3746]. Last Accessed: 20/10/2011. Ministry of Information (2008) Kexue Fazhanguan Xuexi Duben (Scientific Development Outlook Study Guide), Beijing: Study Press. Ministry of Information (2009) 50ge “Weishenme”: Weihu Guojia Tongyi, Fandui Minzu Fenlie, Jiaqiang Minzu Tuanjie Duben (The 50 Whys: Protecting National Unification, Opposing Ethnic Separatism, Strengthening Ethnic Unity Study Book), Wulumuqi: Xinjiang Education Press. Ministry of Information, Theoretical Department (2009a) Liuge Weishenme: Dui Jige Zhongda Wenti de Huida (The Six Whys: Answers to Some Huge Questions), Beijing: Study Press. Ministry of Information, Theoretical Department (2009b) Lilun Redian Mianduimian (Face to Face Hot Theory Topics), Beijing: People’s Publishing Press. Ministry of Information, Education Department (2009c) Minzu Tuanjie Jiaoyu: Tongsu Duben (Ethnic Unity Education: Basic Study Guide), Beijing: Study Press. Mitter, Rana (2004) A Bitter Revolution: China’s Struggle with the Modern World. Oxford: Oxford University Press. Moore, Malcolm (2009) ‘Xinjiang Protests: Han Chinese Demand Better Security After Syringe Attacks in Urumqi’ (Available at: http://www.telegraph.co.uk/news/worldnews/asia/china/6131874/Xinjiang-protests-Han- Chinese-demand-better-security-after-syringe-attacks-in-Urumqi.html), The Daily Telegraph, 3 September [Last Accessed: 22/06/12]. Morgan, Lewis (2005) Ancient Society. New York: Adamant Media Corporation. Morrison, John (2003) Painting the Nation: Identity and Nationalism in Scottish Painting, 1800- 1920. Edinburgh: Edinburgh University Press. Mullaney, Thomas S (2011) Coming To Terms with the Nation: Ethnic Classification in Modern China. Berkeley: University of California Press. Naughton, Barry (1995) Growing Out of the Plan: Chinese Economic Reform 1978-1993. Cambridge: Cambridge University Press. Naughton, Barry (2004) ‘The Western Development Program’ in Naughton, B and Yang, D (eds) Holding China Together: Integration in the Post-Deng Era, pp.253-296. Cambridge: Cambridge University Press.

271

Naughton, Barry and Yang, Dali (2004) ‘Holding China Together: Introduction’ in Naughton, B and Yang, D (eds) Holding China Together: Integration in the Post-Deng Era, pp.253-296. Cambridge: Cambridge University Press. New Dominion (2010) ‘Ilham Tohti: Blacklists that Prohibit Leaving the Country Openly Trample the Law’ (Available at: http://www.thenewdominion.net/?p=1543), 27 April [Last Accessed: 07/08/12]. New York Times (NYT) (1999) ‘In a Far-Flung Conrner of China, A Folk Star’ (Available at: http://www.nytimes.com/1999/02/07/arts/music-in-a-far-flung-corner-of-china-a-folk- star.html?pagewanted=all&src=pm), 17 February [Last Accessed: 08/04/12]. New York Times (NYT) (2007) ‘Chinese Leave Guantanamo for Albanian Limbo’ (Available at: http://www.nytimes.com/2007/06/10/world/europe/10resettle.html?pagewanted=all), 10 June [Last Accessed: 01/09/12]. New York Times (NYT) (2009) ‘To Protect an Ancient City, China Moves to Raze It’ (Available at: http://www.nytimes.com/2009/05/28/world/asia/28kashgar.html?pagewanted=all), 27 May [Last Accessed: 08/04/12]. New York Times (NYT) (2009b) ‘Without Explanation, China Releases 3 Activists’ (Available at: http://www.nytimes.com/2009/08/24/world/asia/24china.html?_r=1&emc=eta1), 23 August [Last Accessed: 07/08/12]. Newby, Laura J (1999) ‘The Chinese Literary Conquest of Xinjiang’, Modern China, 25, pp.451- 474. Newby, Laura J (2005) The Empire and the Khanate, Leiden: Brill. Newby, Laura J (2007) “‘Us and Them’ in Eighteenth and Nineteenth Century Xinjiang” in Bellér- Hann, I et al (2007) (eds) Situating the Uyghurs Between China and Central Asia, Aldershot: Ashgate. Nye, Joseph S Jr. (2004) Soft Power: The Means to Success in World Politics. New York: Perseus Books. Nyíri, Pál (2006) The Yellow Man's Burden: Chinese Migrants on a Civilizing Mission, The China Journal, 56, pp.83-106. Pan, Jiao (ed) (2008) Zhongguo Shehui Wenhua Renleixue / Minzuxue Bainian Wenxuan (‘Selected Works of 100 Years of Chinese Social and Cultural Anthropology / Minzu Studies’). Beijing: Zhishi Chanquan Chubanshe. Pan, Jiao (2008b) ‘Zuqun Jiqi Xiangguan Gainian zai Xifang de Liubian’ (‘Ethnicity and Related Concepts in the Later Developments of the West’) in Pan, Jiao (ed) (2008) Zhongguo Shehui Wenhua Renleixue / Minzuxue Bainian Wenxuan (‘Selected Works of 100 Years of Chinese Social and Cultural Anthropology / Minzu Studies’), pp.83-92. Beijing: Zhishi Chanquan Chubanshe.

272

Pan, Wei (2007) ‘The Chinese Model of Development’ (Available at: http://fpc.org.uk/fsblob/888.pdf), London: Houses of Parliament, sponsored by the Foreign Policy Centre, 11 October [Last Accessed: 23/10/12]. Pan, Zhiping (1996a) ‘Wulumuqi Dutong’ (‘Urumqi Military Governor’) in Yu, Taishan et al (eds) Xinjiang Gezu Lishi Wenhua Cidian (Dictionary of History and Culture of Every Xinjiang Minzu), Beijing: Zhonghua Shuju Chuban, p.51. Pan, Zhiping (1996b) ‘Hancheng’ (‘Han City’) in Yu, Taishan et al (eds) Xinjiang Gezu Lishi Wenhua Cidian (Dictionary of History and Culture of Every Xinjiang Minzu), Beijing: Zhonghua Shuju Chuban, p.86. Pan, Zhiping (2008) “Dongtu” de Lishi yu Xianzhuang (The History and Present Situation of “East Turkestan”), Beijing: Minzu Chubanshe. PBS (2009) ‘Tensions Remain High in China Following Deadly Riots’ (Available at: http://www.pbs.org/newshour/bb/asia/july-dec09/china2_07-07.html), 7 July [Last Accessed: 24/01/13]. PEN (2009) [http://www.englishpen.org/] (last accessed: 01/12/11). People’s Daily (2008) ‘Top Ten Catch Phrases in China Since 1978 Reform’ (http://english.peopledaily.com.cn/90001/90776/90785/6553124.html), 15/12/2008 [Last Accessed: 01/12/11]. People’s Daily (2011) ‘Xinjiang AIDS Attack Rumour Denied’ (Available at: http://english.peopledaily.com.cn/90882/7648910.html), 18 November [Last Accessed: 21/06/12]. Perdue, Peter (2005) China Marches West: The Qing Conquest of Central Eurasia, Cambridge MA: Harvard University Press. Phillips, Nicola (ed) (2005) Globalizing International Political Economy. Basingstoke: Palgrave Macmillan. Pieke, Frank (2000) 'Serendipity: Reflections on Fieldwork in China' in Dresch, Paul et al (eds) Anthropologists in a Wider World: Essays on Field Research, New York: Bergham Books. Polanyi, K. (2010) The Great Transformation: The Political and Economic Origins of Our Time. Boston, Massachusetts: Beacon Press. Prakash, Gyan (1995) ‘Orientalism Now’, History and Theory, 34/3, pp.199-212. Provisional Procedures for the Evaluation of the Registration of Exemplars of National Level Intangible Cultural Heritage (2005c). Pye, Lucian (1990) ‘Erratic State, Frustrated Society’, Foreign Affairs, 69/4, pp.56-74. Quanguo Renmin Daibiao Dahui Changwu Weiyuanhui (NPC) (2001) Zhonghua Renmin Gonghheguo GuojiaTongyong Yuyan Wenzifa (The People’s Republic of China National Language and Script Law) (available at: http://news.xinhuanet.com/legal/2003- 01/21/content_699566.htm) [Last Accessed: 06/04/09).

273

Radio Free Asia (RFA) (2005) ‘I Want to Make My Fights International: Rebiya Kadeer’ (Available at: http://www.rfa.org/english/background/kadeer_text-20050404.html), 4th April [Last Accessed: 17/09/12]. Radio Free Asia (2009) ‘Witnesses Describe Two-Way Violence’ (Available at: http://www.rfa.org/english/news/uyghur/witnesses-07172009121028.html), 17 July [Last Accessed: 15/06/12]. Radio Free Asia (2009b) ‘Bloggers Report Beatings, Death’ (Available at: http://www.rfa.org/english/news/uyghur/online-09082009133813.html), 9 September [Last Accessed: 22/06/12]. Radio Free Asia (2009c) ‘Uyghur Scholar Calls for Jobs’ (Available at: http://www.rfa.org/english/news/uyghur/tohti-03062009130647.html), 6 March [Last Accessed: 07/08/2012]. Radio Free Asia (2009d) ‘Outspoken Economist Presumed Held’ (Available at: http://www.rfa.org/english/news/uyghur/Tohti-07082009151608.html), 8 July [Last Accessed: 07/08/12]. Radio Free Asia (2011) ‘Laid-Off Profs Reject Deal’ (Available at: http://www.rfa.org/english/news/uyghur/deal-09272011172719.html), 27/09/2011, [Last Accessed: 02/04/12]. Radio Free Asia (2011b) ‘Deported Uyghurs Face Terrorism Charges’ (Available at: http://www.rfa.org/english/news/uyghur/charges-06142011163646.html), 14 June [Last Accessed: 07/08/12]. Radio Free Asia (RFA) (2012a) ‘Search For a Missing Son’ (Available at: http://www.rfa.org/english/news/uyghur/search-05112012123934.html), 11 May [Last Accessed: 16/06/12]. Radio Free Asia (RFA) (2012b) ‘Party Member Among Missing’ (Available at: http://www.rfa.org/english/news/uyghur/member-06142012174311.html), 14 June [Last Accessed: 16/06/12]. Radio Free Asia (RFA) (2012c) ‘Buses Asked to Eject Uyghur Petitioners’ (Available at: http://www.rfa.org/english/news/uyghur/bus-04222012170211.html), 22 April [Last Accessed: 16/06/12]. Radio Free Asia (2012d) ‘Police Witness to Riots Dies’ (Available at: http://www.rfa.org/english/news/uyghur/policeman-06182012155524.html), 20 June [Last Accessed: 25/06/12]. Radio Free Asia (2012e) ‘Student Battles Travel Ban’ (Available at: http://www.rfa.org/english/news/uyghur/travel-12202012143138.html), 20 December [Last Accessed: 13/03/13]. RAND Corporation, National Security Research Division (2007) The Beginner’s Guide to Nation- Building, Santa Monica: RAND.

274

Rapport, Nigel (2002) ‘The Truth of Movement, the Truth as Movement: 'Post-cultural anthropology' and Narrational Identity’ in Amit, Vered and Rapport, Nigel (eds) The Trouble With Community: Anthropological Reflections on Movement, Identity, and Collectivity, London: Pluto Press. Rayila, M (2011) ‘The Pain of a Nation: The Invisibility of Uyghurs in ’, The Equal Rights Review, 6. Ren, Zhongping (2009) ‘Gaibian Lishi de “Beijing Shijian”’ (‘Beijing Time: Changing History’), Chen Bao (‘Morning News’), 28 September, p.A2. Renmin Huabao (2005) ‘The Wonderful Women of Xinjiang’ (Available at: http://www.rmhb.com.cn/chpic/htdocs/english/content/200505/10-1.htm), 1 May [Last Accessed: 28/06/12]. Roberts, Sean R (2004) ‘A “Land of Borderlands”: Implications of Xinjiang’s Trans-border Interaction’, in Starr, S.F (ed) Xinjiang: China’s Muslim Borderland, pp.216-240. New York: ME Sharpe. Roberts, Sean R (2009) ‘Imagining Uyghurstan: Re-Evaluating the Birth of the Modern Uyghur Nation’, Central Asian Survey, 28/4, pp.361-381. Roberts, Sean R (2009b) ‘Ethnic Clashes in Xinjiang: Uighurs vs. Han Chinese’, The Washington Post (Available at: http://www.washingtonpost.com/wp- dyn/content/discussion/2009/07/07/DI2009070701491.html), 8 July [Last Accessed: 11/06/12]. Roberts, Sean R (2012) ‘Imaginary Terrorism? The Global War on Terror and the Narrative of the Uyghur Terrorist Threat’, Ponars Eurasia Working Paper. Roe, Paul (1999) ‘The Intrastate Security Dilemma: Ethnic Conflict as a “Tragedy”?’, Journal of Peace Research, 36/2, pp.183-202. Ruan, Xihu (2004) ‘Minzu Haishi “Zuqun” – Yi “Ethnic Group” Yici de Hanyi’ (‘Minzu or “Zuqun” – The Implications of Translating “Ethnic Group” ’), Guanxi Minzu Xueyuan Xuebao, 5. Rudelson, Justin (1997) Oasis Identities, New York: Columbia University Press. Sabah (2012) ‘A Flood of Love in Urumqi’ (Available at: http://english.sabah.com.tr/National/2012/04/09/a-flood-of-love-in-urumqi), 9 April [Last Accessed: 25/06/12]. Saich, Tony (2004) Governance and , Basingstoke: Palgrave. Said, Edward W (1993) Culture and Imperialism. New York: Alfred Knopf. Said, Edward W (1997) Covering Islam. London: Vintage. Said, Edward W (2003) Orientalism. London: Penguin Books. Sautman, Barry (1997) ‘Myths of Descent, , and Ethnic Minorities in the People’s Republic of China’ in Dikotter, F (ed) The Construction of Racial Identities in China and Japan, pp.75-95. London: Hurst & Co. Sautman, Barry (1998) ‘Preferential Policies for Ethnic Minorities in China: The Case of Xinjiang’,

275

Nationalism and Ethnic Politics, 4/1, pp.86-118. Schein, Louisa (1997) ‘Gender and Internal orientalism in China”, Modern China, 23/1, pp.69-98. Schein, Louisa (2000) Minority Rules, London: Duke University. Schichor, Yitzhak (2009) ‘The Uyghur Predicament in Sino-Turkish Relations’, CCS Seminar, University of Manchester, 18/2/2009. Schluessel, Eric (2007) ‘“Bilingual” Education and Discontent in Xinjiang, Central Asian Survey, 26/2, pp.251-277. Schluessel, Eric (2009) 'History, Identity, and Mother-Tongue Education in Xinjiang', Central Asian Survey, 29/4, pp.383-402. Scott, James C (1985) Weapons of the Weak: Everyday Forms of Peasant Resistance. New Haven: Yale University Press. Shanghai Daily (2012) ‘China Mobile Phone Users Exceed One Billion’ (Available at: http://www.shanghaidaily.com/nsp/Business/2012/03/30/China%2Bmobile%2Bphone%2 Busers%2Bexceed%2B1%2Bbillion/), 30 March [Last Accessed: 17/01/13]. Shapiro, Michael J (1988) The Politics of Representation: Writing Practices in Biography, Photography, and Policy Analysis. Madison: University of Wisconsin Press. Shapiro, Michael J (1989) 'Textualizing Global Politics' in Shapiro, Michael J and Der Derian, James (eds) International/Intertextual Relations: Postmodern Readings of World Politics, Lexington Books: New York, p.11-22. Sheehan, Jackie (2009) ‘China’s Dangerous Dilemma in Xinjiang’ (Available at: http://www.nottingham.ac.uk/cpi/publications/commentaries-reports/2009/sheehan- xinjiang-07-09-2009.aspx), China Policy Institute, 9 July [Last Accessed: 11/07/12]. Shen, Jianhua (2009) ‘Gaoxiao Wending Xingshi Fenxi Ji Duice Sikao’, Xinjiang Shifan Daxue Xuebao (Xinjiang Normal University Journal), 30/4, pp.12-14. Shichor, Yitzhak (2009) ‘Ethno-Diplomacy: the Uyghur Hitch in Sino-Turkish Relations’, East- West Center, Policy Studies, 53. Shih, Chih-Yu (2002) Negotiating Ethnicity in China: Citizenship as a Response to the State, Routledge: London. Shijian Bianji Bu (1965) ‘Bixu Bawo Minzu Wenti de Jieji Shizhi’ (‘We Must Grasp the the Class Essence of the Minzu Problem’) in Pan, Jiao (ed) (2008) Zhongguo Shehui Wenhua Renleixue / Minzuxue Bainian Wenxuan (‘Selected Works of 100 Years of Chinese Social and Cultural Anthropology / Minzu Studies’). Beijing: Zhishi Chanquan Chubanshe, pp.213-221. Shue, Vivienne (2004) 'Legitimacy Crisis in China?' in Gries, Peter Hays and Stanley Rosen (eds) State and Society in 21st-century China. New York: Routledge Curzon, pp. 24-49. Shule County Government (2009) ‘Zhulao Fangxian Xiandi Shentou, Changzhi jiuan gujiben Shulexian Wucuo bing Jucuo Quebao Yishi Xingtai Lingyu Fanfenlie Douzheng Qude Shixiao ’ ('A firm line of defence to resist infiltration, ensure long-term peace and order,

276

and secure a foundation. Five measures of Shule County to be carried out in order to ensure success in the ideological fight against anti-separatism') (Available at: http://www.360doc.com/content/11/0512/16/6133644_116221536.shtml) [Last Accessed: 29/08/12]. Smagulova, Juldyz (2009) 'Language Policies of Kazakhization and Their Influence on Language Attitudes and Use', International Journal of Bilingual Education and Bilingualism, 11 (3&4), p.440-475. Smith, Anthony D (1991) The Ethnic Origins of Nations. Oxford: Blackwell. Smith, Anthony D (1995) ‘Gastronomy or Geology? The Role of Nationalism in the Reconstruction of Nations’, Nations and Nationalism, 1, pp.3-23. Smith Finley, Joanne, N (1999) ‘Changing Uyghur Identities in Xinjiang in the 1990s’, Unpublished PhD Thesis. Smith Finley, Joanne N (2000) ‘Four Generations of Uyghurs: The Shift Towards Ethno-Political Ideologies Among Xinjiang’s Youth’, Inner Asia, 2, pp.195-224. Smith Finley, Joanne N (2002) ‘Making Culture Matter: Symbolic, Spatial, and Social Boundaries Between Uyghurs and Han Chinese’, Asian Ethnicity, 3/2, pp.153-174. Smith Finley, Joanne N (2006) 'Maintaining Margins: The Politics of Ethnographic Fieldwork in Chinese Central Asia', The China Journal, 56, p.131-147. Smith Finley, Joanne N (2007) ' 'Ethnic Anomaly' or Modern Uyghur Survivor? A Case Study of the Minkaohan Hybrid Identity in Xinjiang' in Bellér-Hann, I et al (2007) (eds) Situating the Uyghurs Between China and Central Asia, Aldershot: Ashgate. Smith Finley, Joanne N (2011) ‘No Rights Without Duties: Minzu Pingdeng in Xinjiang Since the 1997 Ghulja Disturbances’, Inner Asia, 13/1, pp.73-96. Smith, Kristie N (2006) ‘Dance for Development’, East-West Center Working Papers, No.25. Smith, Neil (1996) The New Urban Frontier: Gentrification and the Revanchist City. London: Routledge. Spegele, Brian (2011) ‘No Foreign Tie Seen in China Attacks’, (Available at: http://online.wsj.com/article/SB10001424053111904491704576574412601080364.html) [Last Accessed: 30/05/2012]. Spivak, Gayatri Chakravorty (1988) ‘Can the Subaltern Speak?’ in Nelson, C and Grossberg, L (eds) Marxism and the Interpretation of Culture. Basingstoke: Macmillan Education, pp.271-313. Stalin, Josef (1913/1945) Marxism and the National Question. Moscow: Foreign Languages Publishing House. Starr, Frederick (2004) ‘Introduction’ in Starr, F (ed) Xinjiang: China’s Muslim Borderland, New York: ME Sharpe, pp.3-24. State Administration of Cultural Heritage (2008) ‘Chinese Section of the Silk Road: Land routes in Henan Province, Province, Gansu Province, Qinghai Province, Ningxia Hui

277

Autonomous Region, and Xinjiang Uygur Autonomous Region; Sea Routes in Ningbo City, Zhejiang Province and Quanzhou City, Fujian Province - from Western-Han Dynasty to Qing Dynasty’ (Available at: http://whc.unesco.org/en/tentativelists/5335/) [Last Accessed: 03/06/09]. State Council (2001a) Circular of the State Council on Policies and Measures Pertaining to the Development of the Western Region (available at: http://english.enorth.com.cn/system/2001/05/30/000002026.shtml) [last accessed: 09/04/09]. State Council (2001b) Circular of the State Council's General Office on the Distribution of "Suggestions on the Implementation of Policies and Measures Pertaining to the Development of the Western Region" (Available at: http://investchina.org.cn/market/hwc/400856.htm), 29 September 29th, [Last Accessed: 09/04/09]. State Council (2002) ‘East Turkestan Forces Cannot Get Away With Impunity’ (Available at: http://english.people.com.cn/200201/21/eng20020121_89078.shtml), 21/1/2002 [Last Accessed: 29/08/12]. State Council (2003) Xinjiang de Lishi yu Fazhan (The History and Development of Xinjiang) (Available at: http://english.gov.cn/official/2005-07/28/content_17948.htm), Beijing: People’s Publishing Press. State Council (2005a) Guanyu Jiaqiang Woguo Feiwuzhi Wenhua Yichan Baohu Gongzuo de Yijian (Suggestions Regarding the Strengthening of the Protection of Our Nation’s Intangible Culture Work) (Available at: http://www.gov.cn/zwgk/2005- 08/15/content_21681.htm) (15/08/2005) (Last Accessed: 30/03/2009). State Council (2005b) Guowuyuan Guanyu Jiaqiang Wenhua Yichan Baohu de Tongzhi (Notification Regarding the Strengthening of Cultural Heritage Protection) (Available at: http://www.law-lib.com/law/law_view.asp?id=129095) [last accessed: 01/04/09]. State Council (2006) Xibu Da Kaifa “Shiyiwu” Guihua (Western Development Project: 11th 5 Year Plan) (http://www.chinawesthr.org/html/xiangmuyanjiuyurencaikaifa/zhengcefagui/20081017/4 0.html) [Last Acccessed: 10/04/09]. State Council (2009a) Zhongguo de Minzu Zhengce yu ge Minzu Gongtong Fanrong Fazhan (China’s Ethnic Minority Policy and the Common Prosperity of all Ethnic Groups), Beijing: People’s Publishing Press. State Council (2009b) Xinjiang de Fazhan yu Jinbu (The Progress and Development of Xinjiang), Beijing: People’s Publishing Press. Statistics Bureau of Xinjiang Uyghur Autonomous (2007) Xinjiang Tongji Nianjian (Xinjiang Statistical Yearbook), Beijing: China Statistics Bureau.

278

Szadziewski, Henry (2011) ‘Commanding the Economy: The Recurring Patterns of Chinese Central Government Development Planning Among Uyghurs in Xinjiang’, Inner Asia, 13/1, pp.97-116. Strange, Susan (1970) ‘International Economics and International Relations: A Case of Mutual Neglect’, International Affairs, 46/2, pp.304-315. Taynen, Jennifer (2006) ‘Interpreters, Arbiters, or Outsiders: The Role of the Minkaohan in Xinjiang Society’, Journal of Muslim Minority Affairs, 26(1), pp.45-62. The Economist (2008) ‘The Second Long March’ (http://www.economist.com/node/12758848), 11/12/2008 [Last Accessed: 01/12/11]. The Guardian (2012) ‘China Steps Up Campaign Against Ramadan in Xinjiang’ (Available at: http://www.guardian.co.uk/world/2012/aug/03/china-restriction-ramadan-xinjiang- uighurs), 3 August [Last Accessed: 23/10/12]. The Independent (2012) ‘12 Children Hurt as China Police Raid Religious School’ (Available at: http://www.independent.co.uk/news/world/asia/12-children-hurt-as-china-police-raid- religious-school-7820183.html), 6 June [Last Accessed: 23/10/12]. Thornton, Patricia (2007) Disciplining the State: Virtue, Violence, and State-Making in Modern China, Cambridge MA: Harvard University Press. Thogersen, Stig and Heimer, Maria (2006) 'Introduction' in Heimer, Maria and Thogersen, Stig (eds) Doing Fieldwork in China, Copenhagen: NIAS Press, pp.1-23. Tianshan.net (2006) ‘Jingdian Tupian: Kuerban Dashu Jiandaole Mao Zhuxi’ (‘Classic Photograph: Uncle Kuerban Meets Chairman Mao’) (http://www.tianshannet.com.cn/pn/content/2006-11/02/content_1316604.htm), 2 November [Last Accessed 22/03/12]. Tilly, Charles (1992) Coercion, Capital, and European States, AD 990-1992. Cambridge MA: Blackwell. Tohti, Ilham (2006) ‘Xinjiang Jingji Fazhan Yu Minzu Guanxi’ (‘Xinjiang’s Economic Development and Minzu Relations’) (Available at: http://www.uighurbiz.net/archives/3639), Uighurbiz, 1 January 2006 [Last Accessed: 10/03/2013]. Tohti, Ilham (2009) ‘Zhongguo de Minzu Zhengce Bu xuyao Fansi Ma?’ (‘China’s Minzu Policies Don’t Need to be Re-Considered?’) (Available at: http://www.uighurbiz.net/archives/189), Uighurbiz, 12 November [Last Accessed: 10/03/13]. Tohti, Ilham (2012) ‘Duoshu Weiwuer Ren Renwei Zhengfu Shi Hanren Liyi de Biaoda’ (‘A Majority of Uyghurs Believe that the Government Represents Han Interests’) (Available at: http://www.uighurbiz.net/archives/6493), Uighurbiz, 3 December [Last Accessed: 10/03/13]. Tohti, Ilham (2012b) ‘Yilihamu Pinglun Weiwuer Daxuesheng Atike Huzhao Shenling Zaoju Shijian’ (‘Ilham Considers the Incident of Uyghur University Student Atika’s Passport

279

Application’) (Available at: http://www.uighurbiz.net/archives/6714), 23 December [Last Accessed: 13/03/13]. Toops, Stanley (1999) 'Tourism and Turpan: The Power of Place in Inner Asia/Outer China', Central Asian Survey, 18 (3), pp.303-318. Tu, Weiming (2007) ‘Toward a Dialogical Civilization’, in Dunhua, Zhao (ed) Dialogue of , Religions, and Civilizations in the Era of Globalization, Cultural Heritage and Contemporary Change Series, 3/25, The Council for Research in Values and Philosophy: Washington DC, pp.11-21. Tuerdi, Yiming (2011) ‘Huijia Jianli’ (‘A Painters Biography’) (Available at: http://tuerdi.com/), [Last Accessed: 06/04/12]. Turner, Frederick Jackson (1986) ‘The Significance of the Frontier in American History’ in Ridge, Martin (ed) Frederick Jackson Turner: Wisconsin’s Historian of the Frontier, Madison: State Historical Society of Wisconsin, pp.1-19. Uighur (2008) ‘Wulumuqi Laixin: Zhongqiu shi Shei de Jieri’ (‘A Letter from Ürümchi: Mid- Autumn Festival is Whose Festival?’) (Available at: http://www.uighurbiz.cn/socity/2008/1003/article_7240.htm), Uighurbiz, 3 October [Access no longer Available]. UNESCO: http://portal.unesco.org/en/ev.php- URL_ID=29008&URL_DO=DO_TOPIC&URL_SECTION=201.html [Last Accessed: 03/06/09]. UNESCO (2003) for the Safeguarding of the Intangible Cultural Heritage, MISC/2003/CLT/CH/14 (available at: http://unesdoc.unesco.org/images/0013/001325/132540e.pdf) [Last Accessed: 03/06/09]. US Department of State, Bureau of East Asian and Pacific Affairs (2011) ‘Background Note: China’ (http://www.state.gov/r/pa/ei/bgn/18902.htm) [Last Accessed: 01/12/11]. US Department of State (2012) ‘Country Reports for Human Rights Practices for 2011: China’ (Available at: http://www.state.gov/j/drl/rls/hrrpt/humanrightsreport/index.htm#wrapper) [Last Accessed: 23/08/12]. Uyghur American Association (UAA) (2012) ‘Reports: Uyghur asylum seekers deported from Cambodia sentenced to life, 17 years in prison’ (Available at: http://uyghuramerican.org/old/articles/6721/1/Reports-Uyghur-asylum-seekers-deported- from-Cambodia-sentenced-to-life-17-years-in-prison/index.html), 26 January [Last Accessed: 25/06/12]. Uyghur Human Rights Project (UHRP) (2011) ‘A City Ruled by Fear and Silence, Urumchi Two Years On’ (Available at: http://www.uhrp.org/uaa-and-uhrp-reports-press-releases/new- uhrp-report-city-ruled-fear-and-silence-urumchi-two-years), [Last Accessed: 16/06/12]. Uyghur Human Rights Project (UHRP) (2012a) ‘New Rules Issued by Hotan Company Discriminate Against Uyghur Women’ (Available at: http://uhrp.org/press-release/new-

280

rules-issued-hotan-company-discriminate-against-uyghur-women.html), June 6 [Last Accessed: 23/10/12]. Uyghur Human Rights Project (UHRP) (2012b) ‘Notice Informs Locals of Mandatory Residence Searches in Hotan Community’ (Available at: http://uhrp.org/press-release/notice- informs-locals-mandatory-residence-searches-hotan-community-police-reserve), 18 June [Last Accessed: 23/10/12]. Uyghur Human Rights Project (UHRP) (2012c) ‘China Launches Campaign to Search Houses of Uighur Muslims’ (Available at: http://uhrp.org/news/china-launches-campaign-search- houses-uighur-muslims), June 21 [Last Accessed: 23/10/12]. Wæver, Ole (1995) ‘Securitization and Desecuritization’ in Lipschutz (ed) On Security, pp.46-87. New York: Columbia University Press. Waite, Edmund (2007) 'The Emergence of Muslim Reformism in Contemporary Xinjiang: Implication for the Uyghurs' Positioning Between a Central Asian and Chinese Context' in Bellér-Hann, I et al (2007) (eds) Situating the Uyghurs Between China and Central Asia, Aldershot: Ashgate. Walker, RBJ (1990) 'Security, Sovereignty, and the Challenge of World Politics', Alternatives, 15/1, pp.3-27. Walker, R.B.J. (1997) ‘The Subject of Security’ in Krause, K and Williams, M (eds) Critical Security Studies. Minneapolis: University of Minnesota Press. pp.61-82 Walsh, Eileen Rose (2005) ‘From Nu Guo to Nu Er Guo: Negotiating Desire in the Land of the Mosuo’, Modern China, 31, pp.448-486. Walsh, Kevin (1992) The Representation of the Past: Museums and Heritage in the Post-Modern World. London: Routledge. Wang, Lequan (2009) ‘Jianjue Daji ge zhong Didui Shili de Shentou Fenlie Pohuai Huodong’ (‘Resolutely Strike Hard Against the Infiltration of all Enemies’ Separatist, Harmful Activities’) (Available at: http://leaders.people.com.cn/GB/9087179.html) [Last Accessed: 31/05/12]. Wang, Lequan (2009b) ‘Wang Lequan Guanyu Muqian Shehui Zhian Xingshi de Dianshi Jianghua’ (‘Wang Lequan’s Televised Speech Regarding the Present Public Order Situation’) in Xinjiang Uyghur Autonomous Region Party Commission Ministry of Information (eds) Jiaqiang Minzu Tuanjie, Weihu Xinjiang Wending: Xuanchuan Jiaoyu Cailiao (Yi) (Strengthening Ethnic Unity, Protecting Xinjiang Stability: Information Education Materials No.1), Wulumuqi: Xinjiang People’s Publishing Press. Wang, Lequan (2009c) ‘Wang Lequan zai Quanqu Wending Gongzuo Dianshi Dianhua Huiyi shang de Jianghua’ (‘Wang Lequan’s Televised Speech at the Regional Stability Work Telephone Conference’) in Xinjiang Uyghur Autonomous Region Party Commission Ministry of Information (eds) Jiaqiang Minzu Tuanjie, Weihu Xinjiang Wending:

281

Xuanchuan Jiaoyu Cailiao (Yi) (Strengthening Ethnic Unity, Protecting Xinjiang Stability: Information Education Materials No.1), Wulumuqi: Xinjiang People’s Publishing Press. Watson, George (2009) ' 'Failed States' on the 'Perilous Frontier': Historical Bases of State Formation in Afghanistan and Central Asia' in Mackerras, Colin and Clarke, Michael (eds) China, Xinjiang, and Central Asia: History, Transition, and Crossborder Interaction into the 21st Century, London: Routledge, p.75-93. Watts, Jonathan (2009) ‘Old Suspicions Magnified Mistrust into Ethnic Riots in Urumqi’ (Available at: http://www.guardian.co.uk/world/2009/jul/10/china-riots-uighurs-han- urumqi), The Guardian, 10 July [Last Accessed: 14/09/12]. Wayne, Martin I (2007) ‘Al Qaeda’s China Problem’ (Available at: http://www.atimes.com/atimes/China/IB27Ad01.html), Asia Times, 27 February [Last Accessed: 15/01/13]. Wayne, Martin I (2008) ‘Inside China’s War on Terrorism’, Journal of Contemporary China, 18/59, pp.249-261. Weber, Cynthia (1998) 'Performative States', Millenium - Journal of International Studies, 27/1, pp.77-95. Weber, Cynthia and Lacy, Mark (2011) ‘Securing By Design’, Review of International Studies, 37/3, pp.1021-1044. Wendt, Alexander (1999) Social Theory of International Politics. Cambridge: Cambridge University Press. Wiemer, Calla (2004) ‘The Economy of Xinjiang’ in Starr, S.F (ed) Xinjiang: China’s Muslim Borderland, pp.163-189. New York: ME Sharpe. Winichakul, Thongchai (1994) Siam Mapped. Honolulu: University Of Hawaii. Wong, Edward (2009) ‘New Protests Reported in Restive Chinese Region’ (Available at: http://www.nytimes.com/2009/09/04/world/asia/04china.html?_r=1), New York Times, 3 September [Last Accessed: 22/06/12]. World Bank (2012) China 2030: Building a Modern, Harmonious, and Creative High-Income Society (Available at: http://www.worldbank.org/en/news/2012/02/27/china-2030- executive-summary), [Last Accessed: 24/08/12]. World Uyghur Congress [http://www.uyghurcongress.org/En/home.asp] (last accessed: 09/03/09). Wright, Sue (2000) Community and Communication: The Role of Language in Nation State Building and European Integration. Clevedon: Multilingual Matters Ltd. Wulumuqi Wanshibao (2009) ‘Benbao jiang Paisong 50,000 tao Guozhu Jinian tie’ (This Paper will Distribute 50,000 National Day Commemoration Stickers), 25 September, B17. Xiao, Li (2006) Zhongguo Guoqing (China’s National Conditions), Beijing: Beijing Languages University.

282

Xinhua (2009) ‘Woguo Feiwuzhi Wenhua Yichan Pucha Qude Jieduanxing Chengguo’ (‘The Fruits of Our Country’s Intangible Heritage Investigation Phase’) (Available at: http://news.xinhuanet.com/newscenter/2009-05/18/content_11396011.htm), 18/05/09 [Last Accessed: 06/04/12]. Xinhua (2009b) ‘Xinjiang Sizhou zhi Lu Bingxue Fengqingjie 12 Yuedi Kaimu’ (‘Xinjiang’s Silk Road Ice and Snow Festival Opens at the Start of December’) (available at: http://news.xinhuanet.com/travel/2009-12/12/content_12634151.htm), 12/12/2009 [Last Accessed: 06/04/12]. Xinhua (2009c) ‘Xinjiang Riot Hits Regional Anti-Terror Nerve’ (Available at: http://news.xinhuanet.com/english/2009-07/18/content_11727782.htm), 18 July [Last Accessed: 11/06/12]. Xinhua (2009d) ‘Innocent Civilians Make up 156 in Urumqi Riot Death Toll’ (Available at: http://news.xinhuanet.com/english/2009-08/05/content_11831350.htm), 5 August [Last Accessed: 11/06/12]. Xinhua (2009e) ‘Many Countries say Xinjiang Riot China’s Internal Affair’ (Available at: http://news.xinhuanet.com/english/2009-07/17/content_11726662.htm), 17 July [Last Accessed: 11/06/12]. Xinhua (2009f) ‘Fresh Chaos Erupts in Urumqi’ (Available at: http://news.xinhuanet.com/english/2009-07/07/content_11666941.htm), 7 July [Last Accessed: 11/06/12]. Xinhua (2009g) ‘Police Arrests 1,434 Suspects in Connection with Xinjiang Riot’ (Available at: http://news.xinhuanet.com/english/2009-07/07/content_11664739.htm), 7 July [Last Accessed: 16/06/12]. Xinhua (2009h) ‘Urumqi Police Detain 253 More Suspects’ (Available at: http://news.xinhuanet.com/english/2009-07/29/content_11793364.htm), 29 July [Last Accessed: 16/06/12]. Xinhua (2009i) ‘Urumqi Police Detain Another 319 in Riot Probe’ (Available at: http://news.xinhuanet.com/english/2009-08/02/content_11814102.htm), 2 August [Last Accessed: 16/06/12]. Xinhua (2009j) ‘Police Detain 718 Suspects in Connection With Urumqi Riot’ (Available at: http://news.xinhuanet.com/english/2009-08/04/content_11825993.htm), 4 August [Last Accessed: 16/06/12]. Xinhua (2009k) ‘Chinese President Back Home After Italy Visit’ (Available at: http://news.xinhuanet.com/english/2009-07/08/content_11674664.htm), 8 July [Last Accessed: 19/06/12]. Xinhua (2009L) ‘Urumqi Quiets Down After Protest Against Syringe Attacks’, (Available at: http://news.xinhuanet.com/english/2009-09/04/content_11992622.htm), 4 September [Last Accessed: 21/06/12.]

283

Xinhua (2009m) ‘Chinese medical experts examine Urumqi syringe attacks victims’ (Available at: http://news.xinhuanet.com/english/2009-09/05/content_12001983.htm), 5 September [Last Accessed: 21/06/12]. Xinhua (2009n) ‘4 prosecuted for endangering public security in Urumqi syringe attacks’ (Available at: http://news.xinhuanet.com/english/2009-09/05/content_12001771.htm), 5 September [Last Accessed: 21/06/12]. Xinhua (2009o) ‘Net Access Being Restored in Xinjiang’ (Available at: http://news.xinhuanet.com/english/2009-12/30/content_12727576.htm), 30 December [Last Accessed: 29/06/12]. Xinhua (2009p) ‘Rumourmonger Held over South China Toy Factory Brawl’ (Available at: http://news.xinhuanet.com/english/2009-06/29/content_11616274.htm), 29 June [Last Accessed: 14/09/12]. Xinhua (2009q) ‘Rebiya Kadeer Lies Again’ (Available at: http://news.xinhuanet.com/english/2009-08/11/content_11865477.htm), 11 August [Last Accessed: 17/09/12]. Xinhua (2011) ‘Factbox: Mid-Autumn Festival and its Traditions’ (Available at: http://news.xinhuanet.com/english2010/culture/2011-09/12/c_131134150.htm), 12 September [Last Accessed: 13/07/12]. Xinhua (2012a) ‘No Mercy for Terrorists: Xinjiang Party Chief’ (Available at: http://news.xinhuanet.com/english/china/2012-03/07/c_131452573.htm), 7 March [Last Accessed: 28/06/12]. Xinhua (2012b) ‘Web Day: China’s Women Enjoy Extra Attention on Women’s Day’ (Available at: http://news.xinhuanet.com/english/china/2012-03/09/c_131455612.htm), 9 March [Last Accessed: 28/06/12]. Xinhua (2012c) ‘Full Text of the Human Rights Record of the US in 2011’ (Available at: http://news.xinhuanet.com/english/china/2012-05/25/c_131611554.htm), 25 May [Last Accessed: 23/08/12]. Xinhua (2012d) ‘Chongqing Mayor Denies High Deficit’ (Available at: http://www.china.org.cn/china/2012-05/05/content_25308694.htm), 5 May [Last Accessed: 24/08/12]. Xinhua (2012e) ‘Xinhua Insight: China Never to Copy a Western System’ (Available at: http://news.xinhuanet.com/english/special/18cpcnc/2012-11/12/c_131968691.htm), 12 November [Last Accessed: 14/01/13]. Xinjiang Dushibao (2010) ‘《Yi Jia Ren》 Zhu Xin Chun’ (‘“One Family” Celebrates a New Spring’) (Available at: http://epaper.xjts.cn/ftp/site1/xjdsb/html/2012-07/12/node_2.htm), 3 February [Last Accessed: 12/07/12], C03. Xinjiang Uyghur Autonomous Regional Government (1993) Xinjiang Weiwuer Zizhiqu Yuyan Wenzi Gongzuo Tiaoli (Xinjiang Autonomous Regional Government: Regulations

284

Regarding Language and Script Work) (available at: http://www.shache.gov.cn/zwgk/ShowArticle.asp?ArticleID=442) [last accessed: 07/05/09]. Xinjiang Uyghur Autonomous Regional Government (XUAR) (1996) Xinjiang Weiwuer Zizhiqu Saochu Wenmang Tiaolie (Xinjiang Uyghur Autonomous Region Government: Regulations Regarding the Eradication of Illiteracy) (Available at: http://gtt.xinjiang.gov.cn/10120/10190/2009/15754.htm) [Last Accessed: 15/01/13]. Xinjiang Uyghur Autonomous Regional Government (2001) Xinjiang Weiwuer Zizhiqu Yiwu Jiaoyu Shishi Banfa (Compulsory Education: Practical Measures) (Available at: http://www3.xjnu.edu.cn/gonghui/zzqzczd/6.htm) [Last Accessed: 13/05/09]. Xinjiang Uyghur Autonomous Regional Government (2001b) Xinjiang Weiwuer Zizhiqu Jiaoyu Dudao Banfa (Xinjiang Uyghur Autonomous Regional Government Education Supervision Measures) (Available at: http://dds.ahedu.gov.cn/show.asp?Article_Id=119) [Last Accessed: 13/05/09]. Xinjiang Uyghur Autonomous Region Government (2005) Xinjiang Weiwuer Zizhiqu Chuangjian Wenming Shequ Guanli Banfa (Xinjiang Autonomous Region: Establishing Civilised Communities Supervision Techniques). Xinjiang Uyghur Autonomous Region Government (2005c) Xinjiang Weiwuer Zizhiquji Feiwuzhi Wenhua Yichan Daibiaozuo Shenbao Pingding Zanxing Banfa (Provisional Procedures for the Evaluation of the Registration of Exemplars of Regional Level Intangible Cultural Heritage), Xinzheng Banfa, 162. Xinjiang Uyghur Autonomous Region Government (2007) Xinjiang Weiwuer Zizhiqu Jiben Puji Jiunian Yiwu Pinggu Yanshou Banfa (Compulsory Nine Year Education Mission: Evaluation and Acceptance Measures) (Available at: http://gtt.xinjiang.gov.cn/10120/10185/2009/15679.htm) [Last Accessed: 13/05/09]. Xinjiang Uyghur Autonomous Regional Government (XUAR) (2010) Xinjiang Weiwuer Zizhiqu Shehui Zhian Zonghe Zhili Tiaolie (Xinjiang Uyghur Autonomous Regional Government Public Order Comprehensive Administration Regulations) (Available at: http://www.law- star.com/cacnew/201001/315051147.htm) [Last Accessed: 08/06/12]. Xinjiang Uyghur Autonomous Regional Government General Office (XUAR General Office) (2010) Xinjiang Lingdao Ganbu Shouce (Xinjiang Leading Cadres Manual). Wulumuqi: Xinjiang Tongjiju. Xinjiang Uyghur Autonomous Regional Government: Department of Information (XUAR Dept of Information) (2009) Xinjiang Fazhan Wending: Liuti (Xinjiang’s Development and Stability: 6 Questions), Xinjiang People’s Press: Wulumuqi. Xinjiang Uyghur Autonomous Regional Government Public Security Department (XUAR Gonganting) (2009) ‘Xinjiang Weiwuer Zizhiqu Gonganting Gonggao’ (‘Xinjiang Uyghur

285

Autonomous Regional Government Public Security Department Announcement’), Chen Bao, 7 September, A1. Xinjiang Uyghur Autonomous Region Party Commission Information Department (XUAR Party Commission Information Department) (eds) (2009) Jiaqiang Minzu Tuanjie, Weihu Xinjiang Wending: Xuanchuan Jiaoyu Cailiao Yi (Strengthening Ethnic Unity, Protecting Xinjiang Stability: Information Education Materials No.1), Wulumuqi: Xinjiang People’s Publishing Press. Xinjiang Uyghur Autonomous Region Party Commission Information Department (XUAR Party Commission Information Department) (2009b), Yi Jiaren (One Family) DVD, Xinjiang Dianzi Yinxiang Chubanshe: Wulumuqi. Xin Kuan (2009) Doulang Yi Jia Ren (Available at: http://hi.baidu.com/ouzhouzhannz/item/0cc93380621956ebe596e041), Baidu Kongjian, 19 August [Last Accessed: 12/07/12]. Xiong, (2005) Hexie Shehuilun (Theses on Harmonious Society), Beijing: Shishi Chubanshe. Ya, Hanzhang (1962) ‘Guanyu Minzu Yici de Yiming Tongyi Wenti’ (‘On the Problem of Unifying Translations of Minzu’) in Pan, Jiao (ed) (2008) Zhongguo Shehui Wenhua Renleixue / Minzuxue Bainian Wenxuan (‘Selected Works of 100 Years of Chinese Social and Cultural Anthropology / Minzu Studies’). Beijing: Zhishi Chanquan Chubanshe, pp.115-126. Ya, Hanzhang (1965) ‘Minzu Wenti de Shizhi shi Jieji Wenti’ (‘The Essence of the Minzu Problem is a Class Problem’) in Pan, Jiao (ed) (2008) Zhongguo Shehui Wenhua Renleixue / Minzuxue Bainian Wenxuan (‘Selected Works of 100 Years of Chinese Social and Cultural Anthropology / Minzu Studies’). Beijing: Zhishi Chanquan Chubanshe, p.222-229. Yang, Faren (2005) Xibu Da Kaifa yu Minzu Wenti (The Western Development Project and the Ethnic Question). Beijing: Renmin Chubanshe. Yang, Dali L (2007) 'Language and the Politics of Identity in the Age of Globalisation' in Yuan, I (ed) Is There a Greater China Identity?, Taibei: Institute of International Relations, pp.159-172. Yasin, Nurmehemmet (2005) Wild Pigeon [available at: http://www.rfa.org/english/uyghur/wild_pigeon-20050627.html and http://www.rfa.org/english/uyghur/wild_pigeon2-20050627.html] (last accessed: 09/03/09). Yee, Herbert (2003) ‘Ethnic Relations in Xinjiang: A Survey of Han-Uyghur Relations in Urumqi’, Journal of Contemporary China, 12 (36), pp.431-452. Yeung, Yue Man and Fa, Jian (2004) Developing China’s West. Hong Kong: Chinese University Press. Youtube (2009) ‘Race Riot in China’ (Available at: http://www.youtube.com/watch?v=GUPHVf5yDM4), 29 June [Last Accessed: 14/09/12].

286

Youtube (2011) ‘Arbitrary Detentions of Uyghurs in Urumchi After July 5, 2009’ (Available at: http://www.youtube.com/watch?v=pRcOZOygmZc), 1 July [Last Accessed: 16/06/12]. Yuval-Davis, Nira (2008) Gender and Nation. Sage: London. Zenn, Jacob (2011) ‘Violence Escalates in China’s Xinjiang Province’, (Available at: http://www.ctc.usma.edu/posts/violence-escalates-in-china%E2%80%99s-xinjiang- province) [Last Accessed: 30/05/2012]. Zhang, Haiyang (2006) ‘Jianlun Zhongguohua yu Hexie Shehui’ (‘On Sinicisation and a Harmonious Society’), Sixiang Zhanxian, 2. Zhang, Weiwei (2012) Zhongguo Chudong (‘China Shock’), Shanghai: Shanghai Renmin Chubanshe. Zhang, Weiwei (2011) ‘An Apt Example of a Civilizational-State’ (Available at: http://www.chinadaily.com.cn/opinion/2011-04/27/content_12401986.htm), The China Daily, 27 April [Last Accessed: 24/01/13]. Zhang, Xintai (2009) Xinjiang Shaoshu Minzu Minsu Minqing Huihua Juan (Xinjiang Shaoshu Minzu Folk Customs and Popular Condition Paintings Book). Wulumuqi: Xinjiang Meishu Sheying Chubanshe. Zhao, Jie (2007) Minzu Hexie Yu Minzu Fazhan (Ethnic Harmony And Ethnic Development), Beijing: Minzu Chubanshe. Zhao, Suisheng (2004) A Nation-State By Construction: Dynamics of Modern Chinese Nationalism, Stanford: Stanford University Press. Zhao, Suisheng (1998) 'A State-Led Nationalism: The Patriotic Education Campaign in Post- Tiananmen China', Communist and Post-Communist Studies, 31/3, pp.287-302. Zheng, Yongnian and Chen, Minjia (2007) China’s Regional Disparity and its Policy Responses. Nottingham: China Policy Institute. Zhongguo Minzu Zongjiao Wang (2012) ‘Di er Dai Minzu Zhengce Tantao’ (‘Exploring The Second Generation of Minzu Policies’) (Available at: http://www.mzb.com.cn/html/folder/292573- 1.htm?utm_source=China+Policy&utm_campaign=0025fe4449- RSS_EMAIL_CAMPAIGN&utm_medium=email) [Last Accessed: 24/08/12]. Zhonghua Renmingongheguo Minzu Diquyu Zizhifa (National Law on Regional Autonomy) (2001) (available at: http://hyconference.edu.cn/chinese/2001/Mar/22466.htm) [last accessed: 09/11/11]. Zhonghua Renmin Gongheguo Minzu Quyu Zizhifa (The PRC National Language and Script Law) (2001) (available at: http://hyconference.edu.cn/chinese/2001/Mar/22466.htm) [last accessed: 09/11/11]. Zhou, Heping (2006) ‘Guanyu Feiwuzhi Wenhua Yichan Baohu de Baxiang Gongzuo’ (Eight Items of Work Regarding the Protection of Intangible Cultural Heritage) in Hao, Sumin (ed) Qiangjiu Baohu Feiwuzhi Wenhua Yichan Xibei ge Minzu zai Xingdong (Rescuing and

287

Protecting Intangible Cultural Heritage: Every Northwestern Ethnic Minority Group in Action), Beijing: Minzu Chubanshe. Zhou, Jie (2006) ‘Zhongguo Xinjiang Weiwuer Mukamu Yishu: Renlei Gongtong de Feiwuzhi Wenhua Caifu’ (‘The Muqam Arts of the Uyghurs, Xinjiang, China: The Common Intangible Cultural Treasures of Humanity’) in Hao, Sumin (ed) Qiangjiu Baohu Feiwuzhi Wenhua Yichan Xibei ge Minzu zai Xingdong (Rescuing and Protecting Intangible Cultural Heritage: Every Northwestern Ethnic Minority Group in Action), Beijing: Minzu Chubanshe, pp.243-249. Zhou, Minglang and Ross, Heidi A. (2004) 'Introduction: The Context of the Theory and Practice of China's Language Policy', in Zhou, Minglang (ed) Language Policy in the People's Republic of China: Theory and Practice Since 1949, pp.1-18. Zhu, Weiqun (2012) ‘Dui Dangqian Minzu Lingyu Wenti de Jidian Sikao’ (‘A Few Thoughts on Current Problems in the Field of Ethnicity’) (Available at: http://www.studytimes.com.cn/2012/02/13/01/01_51.htm), Xuexi Shibao, 13 February [Last Accessed: 11/07/12].

288